You are on page 1of 239

July 2009

Terje Gjengedal, ELKRAFT


Master of Science in Energy and Environment
Submission date:
Supervisor:
Norwegian University of Science and Technology
Department of Electric Power Engineering
Offshore Wind Farm Layouts
Performance Comparison for a 540 MW Offshore Wind Farm
Thomas Haugsten Hansen
Problem Description
With the recently accepted goals set by the European Union, stating that 20 % of all energy
production shall come from renewable energy sources in 2020, and the Norwegian governments
ambitious goal of no net CO2 emission in 2050, the development of wind power production is a
more current issue than ever. Especially large scale offshore wind power will require attention to
new focus areas. The wind may be more stable offshore, but there will be less geographical
smoothing effect, so wind variations will still be a key issue. Power transmission, grid connection
and the internal grid structure represent other main challenges for realization of large scale wind
power, and especially for offshore wind farms.
Off the Norwegian coastline, the large water depths of the Norwegian trench make it impossible to
utilize todays technology, where the wind turbines are mounted to the seabed. To reach shallow
enough areas, the wind turbines must be placed at a minimum distance of approximately 100 km
from shore. At these distances, AC cable connection to shore is challenging, due to the high
capacitance of the cables, and HVDC technology may be a competitive option.
Presently, there is also a debate among the offshore wind community regarding the value of
redundancy required in the offshore grid to maximize the energy yield, and the impact it may
impose on the capital costs of the wind farm. The major concern is related to the cost of
supplementary subsea cabling, either in terms of extra length or higher ratings, versus the value
of the decreased losses during normal operation and contingency operation due to the redundancy
provided to the wind farm. For small wind farms, the power loss due to a fault is relatively small,
and redundancy has not been considered a profitable option. With planned wind farms in the 1GW
range, the power and, in consequence, income lost due to a fault may be large enough for
redundancy to be profitable.
With this as the background, this thesis will focus on the effect of the offshore wind farm layout on
the wind farms performance. Different options for the design of the offshore grid and the
transmission to shore will be studied. The wind farm investigated will have an installed capacity of
540 MW, and will be modeled in detail, meaning that all turbines are modeled separately.
Assignment given: 15. January 2009
Supervisor: Terje Gjengedal, ELKRAFT
Preface
This master thesis has been written for the department of Electric Power Engineering at the Norwegian
university of technology and science, NTNU, in cooperation with Statkraft. As I have chosen to spend the
last year of my studies abroad, the work has been carried out at the Royal Institute of Technology, KTH,
in Stockholm, Sweden.

Being abroad has given certain challenges that I would not have experienced if I would have stayed in
Trondheim. Nobody at the department at KTH knew how to use the simulation program PSS/E, which has
been used in this thesis. Also, when it became clear that the version of the program did not function
correctly, I could not just go to the computer engineer at the department to get a new version, since I had
to get it from NTNU. This was eventually solved by having Kurt Salmi from NTNU logging on to my
computer via remote desktop control. Thank you for reading and answering the large amount of mails I
sent you, and for installing the right version of PSS/E on my computer.

Even though they could not help me with the use of PSS/E, I would like to thank the people at the division
of electric power systems at KTH for helping me with other, more technical, issues in relations to my
thesis work. A special thanks to Katherine Elkington for lending me her licentiate thesis, where the
dynamics of the doubly fed induction generator is well explained.

I would especially like to thank Knut Magnus Sommerfelt at Norconsult for acting as the de facto
supervisor of this master thesis. He helped me getting started with PSS/E, and has spent considerable time
helping me get past the rookie mistakes and unexplainable bugs in PSS/E, discussing the results that were
hard to understand and giving me good advice during the six months I have worked on this thesis. He
made me realize that I needed to use the ABB model for HVDC Light, as the VSC HVDC model that
comes with PSS/E does not represent HVDC Light in a good matter when it comes to dynamic
simulations. Also, the VSC HVDC model gave dynamic simulations that did not converged, making
PSS/E crash when I tried to run dynamic simulations.

I would also like to thank Magnus Gustafsson at Statnett for valuable comments, Leif Warland at Sintef
for helping me with the troubleshooting related to PSS/E, and of course Professor Terje Gjengedal for
being the supervisor of the thesis from NTNU and Statkraft.

Finally, to Cecilie, I send the biggest thanks of all for all the support and encouragement!

Thomas Haugsten Hansen
July 19, 2009





Abstract
This master thesis has been written at the Department of Electric Power Engineering at the Norwegian
University of Science and Technology. The work has been carried out at the Royal Institute of Technology
in Stockholm, where the author spent the last year of his studies as an exchange student.

In the thesis, six different designs of the electrical grid of a 540 MW offshore wind farm, placed 100km
off the Norwegian coast, have been studied and compared. At this distance, AC cable transmission might
be difficult because of the reactive power production in the cables. Taking this into consideration, two
options for the transmission system to shore have been studied. In addition to the AC cable transmission,
voltage source converter based HVDC transmission, in the form of HVDC Light, has been studied, giving
a total of 12 models.

The main scope of the thesis was to study the load flow situation and power system performance of the
different offshore wind farm layouts. Two load flow cases were run for each model; the first studying the
model when the active power transmission to shore was maximized, the second studying the model under
a contingency situation. The reliability of the six designs was compared by calculating the expected
number of cable failures during the life time of the wind farm for each design, and what consequence the
disconnection of any cable would have on the power losses. In order to study the effect of the offshore
grid design and transmission system design on the offshore power system stability, dynamic simulations
have also been executed, and the voltage response and rotor speed response following a fault have been
studied.

All simulations have been executed in version 31 of the program PSS/E. The wind farm was modeled full
scale, consisting of 108 wind turbines rated at 5MW. The wind turbines were modeled as doubly fed
induction generators, using the generic wind model that comes with the program.

The load flow simulations showed that an AC cable connection to shore gave lower total system losses
than a DC connection for all designs. The lowest losses were found at the n-sided ring design in the
AC/AC system, and the highest losses were found for the star design in the AC/DC system. These losses
were 2.33% and 8.19% of the total installed capacity, respectively.

In the dynamic simulations, a three phase short circuit fault, lasting 150ms, was applied at three different
places in the system. The simulations showed that except from at the wind turbines that were islanded as a
result of a fault, all dynamic responses were stable. The HVDC Light transmission to shore gave the
highest voltage drops and the lowest voltage peaks offshore. Also, the maximum speed deviation was
found to be larger when using HVDC Light transmission compared to using AC cables, with two
exceptions; the radial and star designs when a fault was applied to the transmission system. A comparison
of the six different grid designs showed that the results were varying. Based on the results in this thesis it
has not been concluded that one of the offshore designs have better dynamic qualities than the other. The
simulation results indicated that this is case specific, and more dependent on where in the offshore grid the
fault occurs rather than the design of the offshore grid.






Table of contents
1 Introduction .............................................................................................................................. 1
2 Wind energy conversion systems ............................................................................................. 2
2.1 Betz law ........................................................................................................................................ 2
2.2 Turbine efficiency factor ............................................................................................................... 3
2.3 Wind turbines ................................................................................................................................. 4
2.3.1 Fixed speed wind turbines ..................................................................................................... 4
2.3.2 Variable speed wind turbines ................................................................................................. 5
2.3.3 Power electronics in wind turbines ........................................................................................ 7
2.3.4 The dynamics of a DFIG ....................................................................................................... 9
3 Wind farm layouts .................................................................................................................. 12
3.1 General wind farm layout ............................................................................................................ 12
3.2 AC layouts for wind farms........................................................................................................... 12
3.2.1 Small AC wind farm ............................................................................................................ 13
3.2.2 Large AC wind farm ............................................................................................................ 13
3.3 AC/DC mixed layouts for wind farms ......................................................................................... 14
3.4 DC layouts for wind farms........................................................................................................... 14
3.4.1 Small DC wind farm ............................................................................................................ 15
3.4.2 Large parallel DC wind farm ............................................................................................... 15
3.4.3 Series DC wind farm............................................................................................................ 16
3.5 Offshore grid design options........................................................................................................ 17
3.5.1 Radial design ....................................................................................................................... 18
3.5.2 Single-sided ring .................................................................................................................. 18
3.5.3 Double-sided ring ................................................................................................................ 19
3.5.4 Star design ........................................................................................................................... 19
3.5.5 Shared ring design ............................................................................................................... 20
3.5.6 N-sided ring design .............................................................................................................. 20
4 Transmission technologies ..................................................................................................... 22
4.1 HVAC transmission ..................................................................................................................... 22
4.1.1 Overhead lines ..................................................................................................................... 22
4.1.2 Cables .................................................................................................................................. 25
4.2 HVDC transmission ..................................................................................................................... 25
4.3 Traditional HVDC ....................................................................................................................... 26


4.4 HVDC Light ................................................................................................................................ 26
4.4.1 Converters ............................................................................................................................ 28
4.4.2 Filters ................................................................................................................................... 29
4.5 HVDC PLUS ............................................................................................................................... 29
5 Power system performance .................................................................................................... 30
5.1 Power system stability ................................................................................................................. 30
5.1.1 Classification of power system stability issues .................................................................... 30
5.1.2 Rotor angle stability ............................................................................................................. 31
5.1.3 Frequency stability ............................................................................................................... 32
5.1.4 Voltage stability ................................................................................................................... 32
5.2 Power system security ................................................................................................................. 33
5.3 Power system reliability............................................................................................................... 33
5.4 Grid codes .................................................................................................................................... 35
6 Simulation models .................................................................................................................. 38
6.1 Physical parameters for the wind farm site .................................................................................. 38
6.2 Static models ................................................................................................................................ 38
6.2.1 Wind turbines ....................................................................................................................... 38
6.2.2 Transformers ........................................................................................................................ 39
6.2.3 Cables .................................................................................................................................. 39
6.2.4 HVDC Light ........................................................................................................................ 40
6.2.5 Onshore generator and SVC ................................................................................................ 42
6.2.6 Fixed shunts ......................................................................................................................... 42
6.3 Dynamic models .......................................................................................................................... 42
6.3.1 Wind turbine model ............................................................................................................. 42
6.3.2 Onshore generator model ..................................................................................................... 43
6.3.3 SVC model ........................................................................................................................... 43
6.3.4 HVDC Light model ............................................................................................................. 43
7 Load flow simulations ............................................................................................................ 44
7.1 Large AC wind farm layouts ....................................................................................................... 44
7.2 AC/DC wind farm layouts ........................................................................................................... 45
7.3 Radial design ............................................................................................................................... 45
7.3.1 AC/AC ................................................................................................................................. 46
7.3.2 AC/DC ................................................................................................................................. 46
7.4 Single sided ring .......................................................................................................................... 47
7.4.1 AC/AC ................................................................................................................................. 47


7.4.2 AC/DC ................................................................................................................................. 48
7.5 Double sided ring ......................................................................................................................... 49
7.5.1 AC/AC ................................................................................................................................. 49
7.5.2 AC/DC ................................................................................................................................. 50
7.6 Shared ring ................................................................................................................................... 51
7.6.1 AC/AC ................................................................................................................................. 51
7.6.2 AC/DC ................................................................................................................................. 52
7.7 N-sided ring (n=4) ........................................................................................................................ 53
7.7.1 AC/AC ................................................................................................................................. 53
7.7.2 AC/DC ................................................................................................................................. 54
7.8 Star design ................................................................................................................................... 56
7.8.1 AC/AC ................................................................................................................................. 56
7.8.2 AC/DC ................................................................................................................................. 57
8 Dynamic simulations .............................................................................................................. 59
8.1 Fault scenario one ........................................................................................................................ 59
8.2 Fault scenario two ........................................................................................................................ 62
8.3 Fault scenario three ...................................................................................................................... 68
9 Discussion .............................................................................................................................. 78
9.1 Load flow simulations ................................................................................................................ 78
9.2 Dynamic simulations ................................................................................................................... 82
10 Conclusion .......................................................................................................................... 85
11 Further work ....................................................................................................................... 87
Bibliography .................................................................................................................................. 88
Appendix 1: Load flow reports ......................................................................................................... I
Appendix 2: Load flow figures .................................................................................................. XXII
Appendix 3: Automatic sequence short circuit reports ......................................................... LXVIII
Appendix 4: Dynamic models .................................................................................................. LXXI
Appendix 5: Dynamic plots .................................................................................................. LXXXII



Table of figures
Figure 1 - Vestas V90 3.0 MW wind turbine ................................................................................................ 2
Figure 2 - Power output as a function of wind speed and pitch angle (left) and turbine efficiency factor as
a function of tip speed ratio and pitch angle (right) ....................................................................................... 4
Figure 3 Type A configuration: Fixed speed control. ................................................................................. 5
Figure 4 - Type B configuration: Limited variable speed control ................................................................. 6
Figure 5 - Type C configuration: Variable speed control with partial scale converter. ................................. 7
Figure 6 - Type D configuration: Variable speed control with full scale frequency converter. .................... 7
Figure 7 Three phase soft starter topology ................................................................................................. 8
Figure 8 - Back to back converter topology .................................................................................................. 9
Figure 9 - Electrical scheme of DFIG ............................................................................................................ 9
Figure 10 - General wind farm layout ......................................................................................................... 12
Figure 11 - Small AC wind farm layout ...................................................................................................... 13
Figure 12 - Large AC wind farm layout ...................................................................................................... 13
Figure 13 - AC/DC wind farm layout .......................................................................................................... 14
Figure 14 - Small DC wind farm layout ...................................................................................................... 15
Figure 15 - Large DC wind farm layout ...................................................................................................... 16
Figure 16 - Series DC wind farm layout based on DC generators ............................................................... 16
Figure 17 - Radial offshore grid .................................................................................................................. 18
Figure 18 - Single sided ring offshore grid .................................................................................................. 18
Figure 19 - Double sided ring offshore grid ................................................................................................ 19
Figure 20 - Star offshore grid ...................................................................................................................... 19
Figure 21 - Shared ring offshore grid .......................................................................................................... 20
Figure 22 - n-sided ring offshore grid .......................................................................................................... 20
Figure 23 - Single phase equivalent circuit of AC transmission line .......................................................... 22
Figure 24 - Equivalent -circuit of a transmission line................................................................................ 23
Figure 25 - Examples of reactive power absorbed by a lossless line as a function of its real load for various
voltage ratings .............................................................................................................................................. 24
Figure 26 - Transmission capacity as a function of the transmission length for AC cables ........................ 25
Figure 27 - Principle of HVDC connection ................................................................................................. 26
Figure 28 - The configuration of a typical bipolar twelve pulse HVDC connection ................................... 26
Figure 29 - HVDC Light configuration ....................................................................................................... 27
Figure 30 - Six pulse transistor based converter configuration ................................................................... 28
Figure 31 HVDC Light AC side voltage signal and the fundamental component of the harmonics ........ 28
Figure 32 - HVDC PLUS output signal (left) and converter configuration (middle and right)................... 29
Figure 33 - Classification of power system stability. .................................................................................. 31
Figure 34 - The failure and reparation process of a component .................................................................. 34
Figure 35 - Fault ride through requirements for power plants connected to a grid point with voltage
>200kV in the Norwegian grid .................................................................................................................... 37
Figure 36 - Reactive power capacity for wind power plants in the Norwegian grid ................................... 37
Figure 37 Capability curve for HVDC converter ..................................................................................... 40
Figure 38 - The HVDC Light model used in PSS/E .................................................................................... 43
Figure 39 - Single line diagram showing the connection to shore, using two AC cables to provide
redundancy ................................................................................................................................................... 44
Figure 40 - Single line diagram showing the connection to shore with one AC cable disconnected .......... 45
Figure 41 - Single line diagram showing the connection to shore, using two HVDC Light connections ... 45
Figure 42 - Radial design with AC transmission normal operating conditions ........................................ 46
Figure 43 - Radial design with AC transmission - worst case operating conditions ................................... 46
Figure 44 - Radial design with DC transmission normal operating conditions ........................................ 46


Figure 45 - Radial design with DC transmission - worst case operating conditions ................................... 47
Figure 46 - Single sided ring, normal operating conditions ......................................................................... 47
Figure 47 - Single sided ring, worst case operating conditions ................................................................... 48
Figure 48 - Single sided ring, normal operating conditions ......................................................................... 48
Figure 49 - Single sided ring, worst case operating conditions ................................................................... 48
Figure 50 - Double sided ring, normal operating conditions ....................................................................... 49
Figure 51 - Double sided ring, worst case operating conditions .................................................................. 49
Figure 52 - Double sided ring, normal operating conditions ....................................................................... 50
Figure 53 - Double sided ring, worst case operating conditions .................................................................. 50
Figure 54 - Shared ring, normal operating conditions ................................................................................. 51
Figure 55 - Shared ring, worst case operating conditions ............................................................................ 52
Figure 56 - Shared ring, normal operating conditions ................................................................................. 52
Figure 57 - Shared ring, worst case operating conditions ............................................................................ 53
Figure 58 - N-sided ring, normal operating conditions................................................................................ 54
Figure 59 - N-sided ring, worst case operating conditions .......................................................................... 54
Figure 60 - N-sided ring, normal operating conditions................................................................................ 55
Figure 61 - N-sided ring, worst case operating conditions .......................................................................... 55
Figure 62 - Star design, normal operating conditions .................................................................................. 56
Figure 63 - Star design, worst case operating conditions ............................................................................ 57
Figure 64 - Star design, normal operating conditions .................................................................................. 57
Figure 65 - Star design, DC transmission, worst case operating conditions ................................................ 58
Figure 66 - Fault scenario one ..................................................................................................................... 59
Figure 67 Fault scenario one. Radial design. P (green), Q (blue), (light blue) and U (purple) of the
wind turbine generator closest to the substation.. ........................................................................................ 60
Figure 68 - Fault scenario one. Maximum speed deviation [pu on system base] and voltage deviation [%
of initial value] ) of the wind turbine generator closest to the substation. ................................................... 61
Figure 69 - Fault scenario two ..................................................................................................................... 62
Figure 70 Fault scenario two. Radial design. P (green), Q (blue), (light blue) and U (purple) of the
wind turbine generator closest to the substation in the same feeder as the fault occurs. ............................. 63
Figure 71 Fault scenario two. Shared ring design. P (green), Q (blue), (light blue) and U (purple) of
the wind turbine generator closest to the substation when the fault occurs in the same feeder. .................. 64
Figure 72 Fault scenario two. Maximum speed deviation [pu on system base] and voltage deviation [% of
initial value] ) of the wind turbine generator closest to the substation when the fault occurs in the same
feeder as the generator is placed. ................................................................................................................. 65
Figure 73 Fault scenario two. Radial design. P (green), Q (blue), (light blue) and U (purple) of the
wind turbine generator closest to the substation when the fault occurs in one of the other feeders.. .......... 66
Figure 74 Maximum speed deviation [pu on system base] and voltage deviation [% of initial value] ) of
the wind turbine generator closest to the substation following fault 2 when the fault occurs in a different
feeder than where the generator is placed. ................................................................................................... 67
Figure 75 - Fault scenario three ................................................................................................................... 68
Figure 76 - Fault scenario three. Radial design. P (green), Q (blue), (light blue) and U (purple) of the
wind turbine generator further away from the offshore substation than the fault point.. ............................ 69
Figure 77 - Fault scenario three. Shared ring design. P (green), Q (blue), (light blue) and U (purple) of
the wind turbine generator further away from the offshore substation than the fault point. ........................ 70
Figure 78 - Fault scenario three. Maximum speed deviation [pu on system base] and voltage deviation [%
of initial value] ) of the wind turbine generator furthest away from the substation when the fault occurs in
the same feeder as the generator is placed. .................................................................................................. 71
Figure 79 Fault scenario three. Radial design. P (green), Q (blue), (light blue) and U (purple) of the
wind turbine generator closer to the offshore substation than the fault point... ........................................... 72


Figure 80 - Fault scenario three. Radial design. P (green), Q (blue), (light blue) and U (purple) of the
wind turbine generator closest to the offshore substation.. .......................................................................... 73
Figure 81 - Fault scenario three. Star design. P (green), Q (blue), (light blue) and U (purple) of the wind
turbine generator at the central bus of the star.. ........................................................................................... 74
Figure 82 - Fault scenario three. Maximum speed deviation [pu on system base] and voltage deviation [%
of initial value] ) of the wind turbine generator bus next to the fault, that is closer to the substation than the
fault. ............................................................................................................................................................. 75
Figure 83 - Fault scenario three. Maximum speed deviation [pu on system base] and voltage deviation [%
of initial value] ) of the wind turbine generator bus closest to the offshore substation in the same feeder as
the fault occurs. ............................................................................................................................................ 76
Figure 84 - Active power losses for the different wind farm layouts.. ........................................................ 78




Table of tables
Table 1 AC wind turbine concepts. ............................................................................................................ 4
Table 2 - Advantages and disadvantages of using power electronics in wind turbine systems ..................... 8
Table 3- Functional area of power production plants . ................................................................................ 36
Table 4 - Cable parameters of the AC cables used in the different wind farm layouts ............................... 39
Table 5 - HVDC Light modules .................................................................................................................. 40
Table 6 Sending and receiving power in the 150kV HVDC Light modules ............................................ 41
Table 7 - Expected number of failures during the lifetime of the wind farm (CFR = 0.08) ........................ 80
Table 8 - Expected number of failures during the lifetime of the wind farm (CFR = 0.20) ........................ 80
Table 9 Extra power losses compared to normal operating conditions, when a cable is lost. .................. 81
Table 10 - Comparison of the speed deviation for dynamic simulations.. ................................................... 82
Table 11- Comparison of the voltage deviation for dynamic simulations.. ................................................ 83

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 1
Introduction

- 1 -

1 Introduction
With the recently accepted goals set by the European Union, stating that 20 % of all energy production
shall come from renewable energy sources in 2020, and the Norwegian governments ambitious goal of
no net CO2 emission in 2050, the development of offshore wind power production is a more current issue
than ever. For Norway, Enova has estimated the physical potential for offshore wind power production to
14 000 TWh.

The wind industry is, together with the solar power industry, the fastest growing energy technology in the
world today. Wind strengths are higher offshore, and the wind turbines are moved away from the visible
surroundings of humans. Nevertheless, outside of the Norwegian coastline, the large water depths of the
Norwegian trench make it impossible to utilize todays technology, where the wind turbines are mounted
to the seabed. This technology can only be used at maximum depths of approximately 60 meters. To reach
areas with water depths not exceeding 60 meters, the turbines must be placed at a minimum distance of
approximately 100 km from the coastline. At these distances, long AC cable connections to shore can lead
to the need of very large and expensive reactive power compensators due to the high capacitance of the
cables. HVDC technology may be utilized instead.

Presently, there is also a debate among the offshore wind community regarding the value of redundancy
required in the offshore grid to maximize the energy yield, and the impact it may impose on the capital
costs of the wind farm. The major concern is related to the cost of supplementary subsea cabling, either in
terms of extra length or higher ratings, versus the value of the decreased losses during normal operation
and contingency operation due to the redundancy provided to the wind farm. For small wind farms, the
power loss due to a fault is relatively small, and redundancy has not been considered a profitable option.
With planned wind farms in the 1GW range, the issue is more current, as the amount of power and, in
consequence, income lost due to a fault may be large enough for redundancy to be profitable.

With this as the background, this thesis will focus on the effect of the offshore wind farm layout on total
power system performance. Different options for the design of the offshore grid shall be compared by
looking at the difference in power losses and ability to provide redundancy. Also, the dynamic behavior of
selected generators in the different grids shall be studied in order to compare the dynamic characteristics
of each design. Two transmission options to shore will be studied. Both of the transmission options use
two parallel connections to shore in order to provide redundancy. The first option is to have two parallel
AC cable connections to shore. The second option is to have two parallel HVDC Light connections. The
difference in power losses and effect of the transmission choice on the dynamic behavior offshore shall be
studied.

All simulations will be executed in the program PSS/E (Power System Simulator for Engineering), version
31. Twelve wind farm models, one for each grid design and transmission choice, shall be developed in this
program. The wind farm will have a total installed power capacity of 540 MW, and will consist of 108
wind turbines, rated at 5MW. The wind farm is modeled in detail, meaning that all wind turbines will be
modeled separately in the model. Realistic parameters for all elements of the wind farm will be chosen.
The HVDC Light model is developed and provided by ABB, while cable data will be gathered from cable
manufacturers. The generator will be modeled as a doubly fed induction generator, using the generic wind
model that comes with PSS/E.

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 2
Wind energy conversion systems

- 2 -

2 Wind energy conversion systems
Figure 1 shows the contents of a Vestas V90 3.0 MW wind turbine. The energy conversion process in the
wind turbine is as follows: When the air mass enters the blades (1) of the turbine, the kinetic energy in the
wind makes the blades rotate. The blades are connected to the blade hub (2) which again is coupled to a
rotational shaft, rotating at the same speed as the blades. This shaft is connected to a gear (3) where the
rotational speed of the shaft is increased to suit the generator (4), which operates at a much higher
rotational speed than the hub. In the generator, the rotating kinetic energy of the shaft is converted to
electric energy. The electric energy is transferred to a high voltage transformer (5) and from the
transformer it is sent down the tower (6) and to the grid.


Figure 1 - Vestas V90 3.0 MW wind turbine (Vestas Wind systems AS)
There are several ways to construct the wind energy conversion system (WECS) of a wind turbine. This
chapter gives an introduction to the physical laws describing the conversion from wind energy to electrical
energy, and it also presents concepts for turbine technology. The standard, state-of-the-art wind turbine
types are presented, along with a DC turbine that may be used in future DC wind farms.
2.1 Betz law
The power contained by an air mass with the air density [kg/m
3
] flowing at a wind speed v [m/s] through
an area A [m
2
] can be calculated as:
3
2
1
Av P
wind
=
2-1
The air density is a function of the height above sea level, and can be expressed as:
RT
gh
e
RT
P
h

=
0
) (
2-2
, where
(h) = air density as a function of altitude [kg/m
3
]
P
0
= standard sea level atmospheric density = 1.225 [kg/m
3
]
R = specific gas constant for air = 287.05 [J/kgK]
1
2
3
4
5
6
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 2
Wind energy conversion systems

- 3 -

T = temperature [K]
g = gravity constant = 9,81 [m/s
2
]
h = altitude above sea level [m]

In a wind turbine, the power in the wind is converted to rotational energy in the rotor of the wind turbine.
This results in a decreased wind speed after the air mass has passed the wind turbine. The wind turbine
cannot utilize all of the power contained by the air mass passing through the rotor area, as the speed of this
air mass then would be equal to zero at the downwind side of the rotor, thus resulting in an accumulation
of air in this area. The theoretical optimum for the utilization of the power in the wind was deduced by
Albert Betz in 1919
wind wind opt
P Av Av P = = 59 , 0
2
1
59 , 0
2
1
27
16
3 3
,

2-3
Hence, if it was possible to extract power without any power losses, it would be possible to utilize
approximately 59% of the available wind power in a wind turbine.
2.2 Turbine efficiency factor
Practically, depending on the efficiency factor of the wind turbine, the extracted electric power from the
wind is in the range of 35 45 %. The efficiency factor of the turbine depends on several factors such as
blade profiles, pitch angle, tip speed ratio and air density.

Considering the losses in the conversion system, the electric power that is extracted from the wind turbine
is often expressed as:

wind opt p e
P C P
,
) , ( =
2-4

, where C
p
(,) is the turbine efficiency factor. is the pitch angle (the angle of the blades towards the
rotational plane). If the pitch angle is low, the blades are almost perpendicular to the wind and if the pitch
angle is high (near 90 degrees) the blades are almost parallel with the hub direction.
is the tip speed ratio; the ratio between the tip speed of the blades and the wind speed, and can be
expressed:
v
R
rotor T

=

2-5
, where
R
rotor
= the turbine rotor diameter [m]

T
= the turbine rotor speed [rad/s]
v = the wind speed [m/s]

Figure 2 shows the mechanical output as a function of the wind speed v and the pitch angle , and the
turbine efficiency factor C
p
as a function of tip speed ratio and the pitch angle for a defined blade
profile. The maximum extracted electric power from the wind is slightly below 45%, peaking at a tip
speed ratio at between eight and nine for a pitch angle at 0, as shown in the right plot.

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 2
Wind energy conversion systems

- 4 -


Figure 2 - Power output as a function of wind speed and pitch angle (left) and turbine efficiency factor as a function of tip
speed ratio and pitch angle (right) (Lundberg, 2006)
2.3 Wind turbines
Wind turbines can operate at either fixed speed or at variable speed. There are also several concepts for
power control in wind turbines. This section will describe the different concepts used for speed- and
power control, and present the standard AC based wind turbine types. Applying speed control as the
criterion, there are four dominating types of wind turbines. Further on, wind turbines can be classified by
considering the power control method used. Taking both of these classification criterions into
consideration, Table 1 indicates the different types of AC wind turbine configurations:

Speed control Power control
Stall Pitch Active stall
Fixed speed Type A Type A0 Type A1 Type A2
Variable speed Type B Type B0 Type B1 Type B2
Type C Type C0 Type C1 Type C2
Type D Type D0 Type D1 Type D2
Table 1 AC wind turbine concepts. The grey zones indicate combinations that are not used in the wind turbine industry
today (Ackermann, 2005).
2.3.1 Fixed speed wind turbines
In fixed speed wind turbines, the speed of the rotor is fixed and determined by the frequency of the supply
grid, the gear ratio and the generator design. A fixed speed wind turbine is characteristically equipped
with an induction generator (wound rotor or squirrel cage). To ensure a smoother grid connection it is
equipped with a soft starter. The soft starter is a simple power electronic component used to decrease the
in-rush current, and thereby limiting the disturbances to the grid. Its topology is shown in Figure 7. The
generator is directly connected to the grid. Since induction generators always draw reactive power from
the grid, a capacitor bank for this type of configuration is used to provide reactive power compensation.
The concept of the type A turbine is shown in Figure 3: The generator is a squirrel cage induction machine
(SCIG).
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 2
Wind energy conversion systems

- 5 -


Figure 3 Type A configuration: Fixed speed control.
The advantages of fixed speed wind turbines are that they are simple, robust and reliable. Also, the
technology is well-proven and the cost of parts is low. Its disadvantages are that the reactive power
consumption is uncontrollable, and the power quality control is limited. Also, the turbine is exposed to
rather large mechanical stresses. Since the wind turbine operates at fixed speed, all fluctuations in the
wind speed is transmitted as torque fluctuations and thus also as fluctuations in the electrical power
delivered to the grid. In the case of weak grids, these fluctuations might lead to voltage fluctuations which
eventually can result in significant line losses. It therefore requires a stiff grid.

To control the power output of a type A wind turbine, there are mainly three concepts that are used; stall
(passive) control, pitch (active) control and active stall control.

Type A0: stall control
The simplest concept is the stall control, where the blades are bolted to the hub at a fixed angle. The
design of the rotor aerodynamics causes the rotor to lose power (stall) when the wind speed exceeds a
predefined level. The advantages of this control method are that it is simple, robust and cheap.
Disadvantages are that it contributes to lower efficiency at low wind speed, there is no assisted startup and
there are variations in the maximum steady-state power due to variations in air density and grid frequency.

Type A1: pitch control
The pitch control allows the blades to be turned out of or into the wind as the power output is too high or
too low. The advantages of this type of control are that it provides good power control, assisted startup
and emergency stop. From an electrical point of view, good power control means that at high wind speeds
the mean output is kept close to the rated power of the generator. The main disadvantage of this method is
that there are higher power fluctuations at high wind speeds. Due to the slow nature of the pitch speed, the
instantaneous power will fluctuate around the rated power output as the turbine is subject to wind gusts.

Type A2: Active stall control
The active stall control controls the stall of the blade actively by pitching the blades. At low wind speeds,
the blades are pitched just like a pitch controlled wind turbine, in order to achieve maximum efficiency. At
high wind speeds, the blades go into a deeper stall by being by being pitched slightly in the opposite
direction of the one the blades of a pitch controlled wind turbine would go into. The advantages of this
method is that it provides a smoother limited power without the high power fluctuations that the pitch-
controlled wind turbines experience, and it is able to compensate variations in air density. The
combination with the pitch mechanism makes it easier to carry out emergency stops and to start up the
wind turbine.
2.3.2 Variable speed wind turbines
Variable speed wind turbines operate at a speed that keeps the tip speed ratio constant at a predefined
value that corresponds to the maximum power coefficient. This is done by adjusting the rotational speed
Gear
SCIG
Soft -
starter
Capacitor
bank
Grid
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 2
Wind energy conversion systems

- 6 -

of the wind turbine as the wind speed varies. Variable speed wind turbines are designed to achieve
maximum aerodynamic efficiency over a wide spectrum of wind speeds. Contrary to the fixed-speed
turbines, the variable speed system provides a generator torque that is almost constant. The power
fluctuations due to variations in the wind strength are absorbed by changes in the generator speed. The
electrical system is more complicated than for fixed speed turbines. Variable wind speed turbines are
typically equipped with an induction or synchronous generator and connected to the grid through a power
converter. The advantages of variable speed wind turbines are an increased energy capture, improved
power quality and reduced mechanical stress on the turbine. The disadvantages are that the more
complicated electrical system leads to losses in the power electronics, the use of more components and
increased cost of equipment.

Due to power limitation considerations, the variable speed concept is only used in combination with a fast
pitch control mechanism, as indicated in Table 1. Variable speed- or active stall controlled wind turbines
are not included here as they potentially lack the capability of fast power reduction which might lead to a
critically high aerodynamic torque. This again might lead to a runaway situation. Underneath the variable
speed concepts that are used today are described:

Type B: Limited variable speed:
This configuration consists of a wound rotor induction generator (WRIG) with a variable rotor resistance.
This additional variable rotor resistance is known as OptiSlip and is a registered trademark of the Danish
manufacturer Vestas. It changes by an optically controlled converter which is mounted on the rotor shaft.
The generator is directly connected to the grid through a transformer, with a capacitor bank and a soft-
starter.

Figure 4 - Type B configuration: Limited variable speed control
The advantage of this configuration is that when the rotor resistance is controlled, the slip is controlled,
and thus the power output is controlled. The circuit topology is quite simple, and the optical coupling
eliminates the need for slip rings that need brushes and maintenance. Compared to the SCIG the speed
range is improved. The disadvantages are that the speed range is typically limited to 0-10% above the
synchronous speed as it is dependent on the variable rotor resistance, that the power control is quite poor
and that the slip power is dissipated in the variable resistances as losses.

Type C: Variable speed with partial scale frequency converter:
This configuration is widely known as the doubly fed induction generator (DFIG) concept. It consists of a
WRIG which is connected to the grid via a transformer. A partial scale frequency converter is used on the
rotor circuit. This provides smoother grid connection and performs the reactive power compensation that
was provided by the soft-starter and capacitor bank respectively.
Gear
Variable resistance
WRIG
Soft -
Starter
Capacitor
bank
Grid
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 2
Wind energy conversion systems

- 7 -


Figure 5 - Type C configuration: Variable speed control with partial scale converter.
This configuration has several advantages. It provides a wide range of dynamic speed control; typically
the speed range comprises synchronous speed -40% to +30%. The DFIG can provide a decoupled active
and reactive power control, and the generator does not necessarily need to be magnetized from the grid.
The smaller frequency converter makes the concept economically desirable. Its main disadvantages are the
use of slip rings that need brushes and maintenance, and that it is poorly protected in the case of grid
faults.

Type D: Variable speed with full scale frequency converter:
In this configuration, the generator can be excited either by a permanent magnet (Permanent magnet
synchronous generator (PMSG)), or electrically (Wound rotor synchronous generator (WRSG) or WRIG).
It is connected to the grid via a full scale frequency converter which performs the reactive power
compensation and the smoother grid connection. The gearbox can be left out for some full-variable speed
wind turbine systems, and in these cases a direct driven multipole generator with relatively large diameter
is used.

Figure 6 - Type D configuration: Variable speed control with full scale frequency converter.
The advantage of this configuration is the full range speed control and that the gearbox can be omitted.
For the permanent magnet generator, the efficiency is higher than in the induction machine as no energy
supply is needed for the excitation. The disadvantage is that to omit the gearbox, a large and heavy
generator is needed, and a full scale power converter that has to handle the full power of the system is
needed. PMSGs may cause problems during startup, synchronization and voltage regulation. The
synchronous operation causes a stiff performance during an external short circuit and if the wind speed is
unsteady. Also, PMSGs are sensitive to temperature, and therefore the rotor temperature must be
supervised and a cooling system is required.
2.3.3 Power electronics in wind turbines
This chapter will give a brief explanation of why power electronics are important in wind turbine systems,
and presents the most commonly used power electronics devises in wind turbine systems. The details of
the operation of the power electronics are beyond the scope of this project, and interested readers are
Gear
Partial scale frequency converter
WRIG
Grid
Gear
PMSG/WRSG/WRIG
Grid
Full scale frequency converter
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 2
Wind energy conversion systems

- 8 -

referred to (Mohan & Undeland, 2006). The two main features of power electronics is that they make it
possible to control the frequency, which is important for the wind turbine.

Power electronic equipment also provides the possibility for wind farms to become active elements in the
power grid, i.e. they can have power plant characteristics. This is important for the grid. The
disadvantages of using power electronics are from the wind turbine perspective the additional power
losses and the increased price of equipment. Concerning the grid side, the power electronics generate high
harmonics. Table 2 lists the advantages and disadvantages of using power electronics in wind turbine
systems:

Power electronics properties Advantages Disadvantages
Controllable frequency
(important for the wind
turbine)
Energy optimal operation
Soft drive train
Load control
Gearless option
Reduced noise
Extra costs
Additional losses
Power plant characteristics
(important for the grid)
Controllable active and reactive
power
Local reactive power source
Improved voltage stability
Improved power quality
- Reduced flicker level
- Filters out low harmonics
- Limited short circuit power
High harmonics
Table 2 - Advantages and disadvantages of using power electronics in wind turbine systems
The two main power electronic devices are the soft starter and the frequency converter, as indicated in
Figure 3to Figure 6. The topology of the soft starter is shown in Figure 7:


Figure 7 Three phase soft starter topology
As shown, it consists of two thyristors connected in antiparallel for each phase. The in-rush current is
controlled by adjusting the firing angle of the thyristors. After the in-rush, the thyristors are bypassed in
order to reduce the power losses.

There are several frequency converter topologies present, but only the most commonly used is described
in this chapter. This is the back-to-back converter, whose topology is shown in Figure 8
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 2
Wind energy conversion systems

- 9 -


Figure 8 - Back to back converter topology
Due to the increasing amount of, and sizes of, wind farms, the wind farms will have to have power plant
characteristics, that is, they have to be able to work as active controllable components in the power
system. The wind farms will have to perform frequency and voltage control, to regulate active and
reactive power and to provide quick responses during transient and dynamic disturbances in the power
system. Without power electronics, there are no solutions as of today that can meet these high demands.
Power electronic devices will therefore play a significant part in large wind farms.
2.3.4 The dynamics of a DFIG
In this thesis, it is chosen to use a model of the DFIG for the simulations. The DFIG is widely used in
wind farms, for instance in the 160 MW Horns rev offshore wind farm (www.hornsrev.dk) where the
VestasV80 turbine is used. A wind power plant consisting of DFIGs can improve the angular behavior of
the power system, but may decrease voltage stability under larger disturbances. DFIGs with power system
stabilizers (PSS) may be used as a positive contribution to power system damping (Elkington, 2009). The
electrical scheme of a DFIG is shown in Figure 9. It resembles the traditional circuit for the induction
machine (Hubert, 2002). The only difference in the electrical scheme is that since the rotor of the DFIG is
also excited, the rotor side is open-circuit instead of being short circuit as in the standard induction
machine model.

Figure 9 - Electrical scheme of DFIG
In the figure, values with the subscript s are located on the stator side of the generator and values with the
subscript r are located on the rotor side. The nomenclature otherwise is as follows:


Grid
Generator
Rectifier DC link Inverter
s j
s s
e V V

=
s
R
s
jX
fe
R
m
X
r s
N N :
r
jX
s
R
r
s
e V
s
V
r j
r r

=
s
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 2
Wind energy conversion systems

- 10 -

V = voltage [V]
s = network voltage phase angle []
s = angular displacement between stator and rotor voltage []
R = resistance []
X = reactance [H]
N = number of windings

R
fe
and X
m
are the iron resistance and magnetic reactance respectively. k s is the slip, which is given by:

s
r
s

= 1
2-6
, where

r
= rotational speed of the rotor [rad/s]

s
= rotational speed of the stator [rad/s]

Using the third order, two-axis model, neglecting the stator resistance, the equations describing the
dynamics of a doubly-fed induction machine without any compensation in the form of voltage regulation
or a PSS are (Elkington, 2009):
) sin( ) sin( ) ( (
1
0 0
0
r r s s s
s
r s
V T V
X
X X
E T
T E
+

=
&

2-7
) (
1
e
r
s
m r
P P
M
=

&
2-8
) ) ( (
1
0 0
0
q
s
q
s
d r s dr s q
V
X
X X
E
X
X
E T V T
T
E

=
&

2-9
) ) ( (
1
0 0
0
d
s
d
s
q r s qr s d
V
X
X X
E
X
X
E T V T
T
E

+ =
&

2-10
s qs qs ds ds s e
P i v i v
X
V E
P = + =

= ) sin(
2-11
r
m
s
X
X
X X
2
=
2-12
r s
r
R
X
T

=
0

2-13
s
n
HS
M

2
=
2-14

, where the subscripts q and d denotes the q and d axis value of the symbols they are attached to. The
subscripts r and s denote that the value in question is connected to the rotor and stator respectively. The
parameter symbols not already accounted for in relation to Figure 9 are:

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 2
Wind energy conversion systems

- 11 -

= angle between transient emf and stator (output) voltage []
E = transient emf [V]
T
0
= open circuit time constant [s]
= rotational speed [rpm]
X = transient reactance [H]
M = inertia coefficient [MVAs
2
]
P
m
= mechanical output power [W]
P
e
= electrical output power [W
P
s
= stator power [W]
H = inertia constant of turbine shaft and generator [s]
S
n
= transient reactance [MVA]

For the background for these equations and further details regarding the modeling of DFIGs, see
(Elkington, 2009).

The simulation tool PSS/E

, which is used for the simulations in this thesis, contains a generic wind
model using this type of generator. The model is described in chapter 6.3.

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 3
Wind farm layouts

- 12 -

3 Wind farm layouts
Depending on how transformers and power electronics are utilized, there are several possibilities for the
layout of a wind farm. This chapter presents and discusses these. The contents of the chapter is based on
(Ackermann, 2005), (Martander, 2002) and (Lundberg, 2006), where several solutions for wind farm
layouts are presented.
3.1 General wind farm layout
The wind farms investigated in this thesis can be represented by the general sketch presented in Figure 10.


Figure 10 - General wind farm layout

Even though the offshore grid in the figure is connected in radials, there are several ways of constructing
the local wind turbine grid (see chapter 3.5 ). As indicated in the figure, a wind farm consists of the
following elements:

Wind turbines
Local wind turbine grid
Collecting point
Transmission system
Wind farm interface to the point of common coupling (PCC)

The wind turbines include generators, power electronics and a voltage adjusting unit in the form of an AC
or DC transformer, as discussed in chapter 2. The local wind turbine grid can be AC or DC and is the grid
connecting the wind turbines together and to the collecting point. The collecting point is an offshore
substation, including the transformer and power electronics used for the respective transmission
technology that is chosen. The transmission system is the connection to shore, where the power is
transmitted to the wind farm grid interface Transmission technologies are further discussed in chapter 4.
The wind farm grid interface adapts the voltage, frequency and reactive power to suit the voltage,
frequency and reactive power demand of the grid in the PCC.
3.2 AC layouts for wind farms
All present wind farms are built using AC technology for the grid and transmission system. In this thesis,
two AC concepts are presented, the small AC wind farm and the large AC wind farm.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 3
Wind farm layouts

- 13 -

3.2.1 Small AC wind farm
In the small AC wind farm, the grid is used not only to connect the wind turbines together, but also to
transfer power from the wind farm to the grid interface. This solution is the most feasible for small wind
farms quite close to shore. The topology is presented in Figure 11

Figure 11 - Small AC wind farm layout
As shown, the local wind farm grid is used both for connecting the wind turbines together and to shore.
The voltage level of such wind farms will mainly depend on the distance to shore and the power rating of
the wind farm. At the Swedish offshore wind warm Utgrunden 1, the voltage level is 22 kV, and at
Bockstigen it is 10,5 kV.
3.2.2 Large AC wind farm
In the large AC wind farm the wind turbines are connected to an offshore substation, where the voltage is
adjusted to minimize transmission losses and reactive power compensation devices are placed. Voltage
levels are dependent on the distance to shore and the power rating of the wind farm. At Horns rev
(www.hornsrev.dk), the offshore grid voltage is 36 kV, while the transmission voltage is 150 kV. The
topology is presented in Figure 12


Figure 12 - Large AC wind farm layout
This solution is suitable for relatively large wind farms that are placed at a considerable distance from
shore. The transfer limit of the AC transmission system is dependent on the distance from shore and is
therefore physically limited by this.

A solution to this could be to decrease the offshore frequency and use an offshore low frequency AC
network, as suggested by for instance (Schtte, Gustavsson, & Strm, 2001) . Low frequency systems are
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 3
Wind farm layouts

- 14 -

used in electrified railway systems, where the frequency ranges from 16.67 Hz to 25 Hz. There are mainly
two advantages of a low frequency AC network:
Firstly, the lowered frequency would increase the transfer capability of the transmission system, as the
capacitive charging current of the cable is significantly reduced when the frequency drops. The
disadvantage of this concept is that transformer sizes would increase, and hence, transformer costs would
increase. Secondly, the low frequency network would allow a simpler design of the offshore wind
turbines. The aerodynamic rotor of a large wind turbine operates at maximum revolutions at 15-20 rpm,
and the lower frequency would therefore allow a smaller gear ratio for turbines with a gearbox, or
decrease the number of poles for wind turbines with direct driven generators. This would lead to lighter
and probably also cheaper turbines.

3.3 AC/DC mixed layouts for wind farms
The mixed AC/DC system consists of wind turbines connected in an offshore AC grid, which is collected
at an offshore station. At this station, the AC power is rectified and transmitted to shore using an HVDC
solution. Onshore, the power is inverted to AC again and fed to the grid. This is shown in Figure 13:


Figure 13 - AC/DC wind farm layout
The DC connection separates the offshore grid from the onshore grid, which makes this type of connection
suitable for wind farms where the distance to the PCC is long or if the AC grid that it is connected to is
weak. The voltage and frequency of the offshore grid are fully controllable from the offshore converter
station, which can be utilized for a collective variable speed system of all the wind turbines in the farm.

An option is, as for the AC layouts, to use a low frequency AC grid on the offshore side. The turbine
advantages described in the previous chapter are still valid, but the AC transmission system is replaced by
a DC system, which removes the need for large and expensive transformers.
3.4 DC layouts for wind farms
When it comes to DC layouts for wind farms, three layouts are presented in this thesis, the small DC wind
farm, the large DC wind farm and the series DC wind farm. The main disadvantages of using DC wind
turbines are that all turbines connected to the same DC/DC transformer will operate at the same speed.
One solution to this challenge is to supply every wind turbine with a variable speed design to have optimal
aerodynamic efficiency of every wind turbine. Another solution is to connect the wind turbines in clusters
to a common converter to save the costs of supplying each wind turbine with a variable speed design.
Finding the best solution is a matter of economical optimization. The best solution is found where the
benefits of reducing the losses by increasing aerodynamic efficiency is equal to the extra costs of
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 3
Wind farm layouts

- 15 -

supplying separate variable speed systems for the next turbine. This is a complicated matter, since there is
a need to predict the price of electrical power for the entire lifetime of the wind turbines to be able to find
the optimal decision. In this chapter, the large DC wind farm denotes the solution where several turbines
are collected at a common DC\DC transformer, and the series DC wind farm denotes a solution where all
wind turbines have a separate variable speed design.
3.4.1 Small DC wind farm
The small DC wind farm has the same electrical system as the small AC wind farm. Each wind turbine is
equipped with a rectifier and the power is sent as DC to the grid interface where it is inverted and fed into
the grid. Figure 14 shows the system topology:


Figure 14 - Small DC wind farm layout
The only difference from the small AC wind farm is that the transformer in the grid interface is replaced
with a DC\DC transformer and an inverter. The advantage of the small DC wind farm versus the large DC
wind farm is that it does not require an offshore substation, the same advantage as the small AC wind farm
has versus the large AC wind farm.
3.4.2 Large parallel DC wind farm
Figure 15 shows the layout of the large DC wind farm. Several wind turbines are connected in clusters to
the first DC\DC transformer step. The power is then sent to the main collecting point, where the DC
voltage is boosted in the second DC\DC transformer and sent to shore. Onshore, the power is rectified and
fed into the grid.

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 3
Wind farm layouts

- 16 -


Figure 15 - Large DC wind farm layout
The difference from the large AC wind farm is that the transformers on and offshore are replaced by
DC/DC converters, and that the wind turbines are connected in parallel to the first offshore converter step.
3.4.3 Series DC wind farm
Figure 16 shows the layout of the series DC wind farm: In this system layout, the wind turbines are
connected in series in order to obtain a voltage suitable for transmission directly. The advantage of this
layout is that it does not require offshore DC-transformers and offshore platforms.

Figure 16 - Series DC wind farm layout based on DC generators
In the topology, n wind turbines are series connected to obtain a voltage suitable for transmission, and m
series-connections are coupled in parallel to obtain the desired power level. The n series-connected
turbines are referred to as stacks. The power is transferred to shore where it is inverted and fed into the
grid at the PCC.

The main drawback is that the DC\DC converters in the wind turbines must have the ability to operate at
high voltage levels The nominal output voltage v
WT,x,y
of a wind turbine can be expressed by:
n
v
v
stack
nom WT
=
,

3-1
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 3
Wind farm layouts

- 17 -

=
=
n
x
y x out y s
P
n
P
1
, , ,
1

3-2
nom WT
y s
stack
n
x
y x out
y stack
v
P
v
P
i
,
,
1
, ,
,
= =

=

3-3
y s
y x out
nom WT y x WT
P
P
v v
,
, ,
, , ,
=
3-4
, where:
x = 1n
y = 1m
v
WT,nom
= Nominal output voltage of one wind turbine [V]
i
stack,y
= Current in stack y [A]
P
out,x,y
= Power output of wind turbine x in stack y [W]
P
s,y
= Mean power production of stack y [A]

Thus, if one or more wind turbines are disconnected, the voltage level of the remaining turbines increases.
Due to this, the wind turbines must be overrated. An overvoltage rating of 1.35 pu is suitable (Lundberg,
2006).
3.5 Offshore grid design options
In this thesis, only an AC offshore grid is studied. The offshore grid denotes the electrical system from the
wind turbines to the (first) offshore substation of the wind farm. Only cable faults are addressed.

The offshore grid can be designed in several ways, depending on the wind farm size and the desired level
of redundancy. Up until today, providing redundancy has not been considered for existing wind farms
since the expected loss of income due to a fault has been assumed to be lower than the costs of the extra
equipment needed to provide redundancy. Nevertheless, as wind farm sizes increase, the amount of energy
(and income) lost during a fault might be high enough for redundancy to become profitable. As a part of
the EU sponsored DOWNVInD (Distant Offshore Wind farms with No Visual Impact in Deepwater)
project, a project group studied and evaluated the offshore grid of offshore wind farms. The content of this
chapter is mainly based on the findings of this project group (Quinonez-Varela, Ault, Anaya-Lara, &
McDonald, 2007). The group found that there are mainly three different conceptual designs that can be
utilized:

Radial design, where all wind turbines are connected to a single series circuit (radial). This has
been used in several small offshore wind farms.
Looped design, where redundancy is provided by the establishment of a looped circuit between
the wind turbines
Star design, where the wind turbines are distributed over several feeders, allowing the use of
lower rated equipment.

The options may all be utilized for both AC and DC solutions. The following chapters present six designs
for the offshore grid in an offshore wind farm, five of which have been suggested by the DOWNVInD
group, while the last one is proposed as a part of this thesis.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 3
Wind farm layouts

- 18 -

3.5.1 Radial design
Figure 17shows the layout of the radial offshore grid. m wind turbines are connected in series to one
feeder and collected at the collector hub. The maximum number of wind turbines that can connect to one
feeder is determined by the cable rating and the generator rating. In practice, this will be case specific as
the geographical constraints will influence on the choice. The main advantage of this design is that it is
inexpensive as the total cable length is smaller than for the other options. Also, it is simple to control, and
it provides the possibility to taper the cable capacity as the distance from the hub increases, since the
amount of power transmitted is smaller further out in each feeder. A stepwise tapering of the feeder might
be an option for long feeders with high total power rating but after conferring with professor Terje
Gjengedal at NTNU (Gjengedal, 2009), this option is not considered here as the extra costs during the
cable installation is considered too high for this to be feasible. The major disadvantage with this design is
the poor reliability provided. Cable or switchgear faults at the hub side of the feeder will lead to the loss of
power from all downstream turbines in the feeder.


Figure 17 - Radial offshore grid
3.5.2 Single-sided ring
Figure 18 shows the single sided ring design. This is a version of the looped design, which addresses some
of the reliability issues of the radial design by providing a redundant path for the power flow within a
feeder. In the single-sided ring design, this additional security comes at the expense of higher cable costs
as the cable length will double. A cable is installed from the outermost turbine in the feeder to the
collector hub. This cable must be able to carry the entire power flow of the feeder in the case of a fault
occurring at a point between the first turbine and the hub, and must therefore have the same power rating
as the original cable. Even though the cable costs increase compared to the radial system, an initial
feasibility study(Quinonez-Varela, Ault, Anaya-Lara, & McDonald, 2007) commissioned by the
DOWNVInD consortium recommended and utilized this design for the 1000 MW wind farm offshore grid
studied.

Figure 18 - Single sided ring offshore grid
m
m
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 3
Wind farm layouts

- 19 -

3.5.3 Double-sided ring
Figure 19shows the single sided ring, which is another version of the looped design. Two feeders are
connected in parallel to provide redundancy. This means that for two feeders, the cable length will only
increase by the distance between the turbines at the end of the feeders. Nevertheless, in the case that a
fault occurs at the cable between the first turbine of a feeder and the collector hub, the full output power of
the wind turbines in the faulted feeder must be diverted through the other, meaning that the cable at the
hub end of the latter needs to be dimensioned for the power output of double the number of wind turbines.
This does not mean that the entire feeder needs to have double power capacity; one can taper the cable
capacity as the distance from the hub increases. This will be an economical issue, where the extra
installation costs must be weighed against the expected value of lost load over the lifetime of the wind
farm. Again, after conferring with professor Terje Gjengedal at NTNU (Gjengedal, 2009), this option is
not considered here as the extra costs during the cable installation are considered too high for this to be
feasible.

Figure 19 - Double sided ring offshore grid
3.5.4 Star design
Figure 20shows the star design concept. The design allows reduced cable ratings and improved security,
since a cable outage will only affect one wind turbine (except in the case where the fault occurs in the
connection to the collector hub). The voltage regulation is also likely to be better in this configuration. The
drawback is the increased expenses due to the longer diagonal cable runs and the short section of the
higher rated connection to the feeder. However, these costs are not likely to be significant, especially in
the presented star shape where nine turbines are coupled together. The major cost of this arrangement is
the more complex switchgear requirement at the central turbine. To provide redundancy in the connection
to the collector hub, two stars could be connected in parallel and the cable rating of the connections to the
collector hub increased. Nevertheless, the cost of this redundancy is considered too high to be a
competitive alternative compared with the current design, as the probability of a cable failure in the short
connection to the collector hub is very low.


Figure 20 - Star offshore grid
m
Connection to collector hub
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 3
Wind farm layouts

- 20 -

3.5.5 Shared ring design
Figure 21 shows the configuration which is given the name shared ring design in this thesis. This
design was developed by the DOWNVInD project group and consists of four feeders connected in parallel
to a redundant cable. The redundant circuit is designed to potentially deliver the full power output of a
failing feeder within the four-feeder arrangement. The probability of two or more feeders failing at the
same time is considered sufficiently small to avoid having higher rated cable capacity for the redundant
cable.

Figure 21 - Shared ring offshore grid
3.5.6 N-sided ring design
As an alternative to the double sided ring, the n-sided ring design (Figure 22) is suggested in this thesis.
The difference from the double sided ring is that instead of simply connecting two feeders in parallel, a
higher number, n feeders, are connected in parallel. The idea is to reduce the high power rating of cables
and equipment which is necessary in the double sided feeder design.

Figure 22 - n-sided ring offshore grid
Defining the numbers of wind turbines in each feeder as m, this means that during steady state, each
feeder j must be able to carry the power S
j
defined as

=
=
m
i
i j
S S
1

3-5

When connecting n feeders in parallel, each feeder j will be dimensioned to carry its own power
production plus the amount of power delivered to the feeder when a fault occurs closest to the collector
hub at one of the other feeders. This power will be divided between the n-1 feeders still in operation,
meaning that each feeder must be dimensioned to handle the worst case fault situation power flow S
j,fault
.
This power can be found by using the basic current division formula as:
m
n
m
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 3
Wind farm layouts

- 21 -

+
+ =
th radial
th
j fault j
Z Z
Z
S S 1
,


Z
radial
is the impedance of the radial closest to the fault, while Z
th
is the thevenin impedance of
the rest of the cables in the n feeders and n-1 cables that connect the feeders in parallel.

3-6
When choosing the amount of feeders to be connected in parallel, one must consider the extra costs of the
parallel connection of one more turbine versus the saved cost of reduced power ratings of the equipment
due to this extra connection. The correct amount of parallel connections is the one where these costs cross
each other.


Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 4
Transmission technologies

- 22 -

4 Transmission technologies
This chapter will present the transmission technologies that are available for the transmission from the
offshore wind farm to the PCC. Also, overhead AC transmission is presented since, from a mathematical
point of view, an underground or subsea cable can be modeled in exactly the same way as an overhead
line. The only difference will be the parameter values. The contents of the chapter are based on the project
work performed the fall of 2008 (Haugsten Hansen, 2008), where a more detailed description of the
technologies is presented.
4.1 HVAC transmission
HVAC transmission is used to connect synchronized AC networks that oscillate at the same frequency and
in phase. Three phase systems are used since they minimize the material needed to transfer the same
amount of power compared to one or two phase systems. In addition, a three phase system applies the
rotating magnetic field needed in the stator of electrical AC-machines, without having to use extra
equipment. Both synchronous machines and induction machines have a simpler configuration than DC-
machines, and thus are cheaper to buy and easier to maintain than DC-machines.
The power losses in an AC-connection are proportional to the square of the current, while the power
transferred is proportional with the square of the voltage. Therefore, it is desirable to have as high voltage
as possible when transferring power.
4.1.1 Overhead lines
When transferring power, an overhead lines impedance consists of the resistance in the line, the
capacitance to the ground, and the line inductance. A single phase equivalent circuit of a transmission line
with distributed parameters is shown in Figure 23:

Figure 23 - Single phase equivalent circuit of AC transmission line
The parameters describing the circuit are:
r = Series resistance per km per phase [/km]
L = series inductance per phase [H/km]
x = L = Series reactance per km per phase [/km].
g = Shunt conductance per km per phase [S/km]
C = Shunt capacitance per phase [F/km]
B = C = Shunt susceptance per unit length per phase [S/km]
l = Line length [km]

The series impedance per km per phase and the shunt admittance per km per phase can be expressed
according to the following equations:
r Is r r r x x x x
g b g b
+
-
+
-
Vs Vg
Ig
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 4
Transmission technologies

- 23 -

jx r z + =
4-1
jb g y + =
4-2
By multiplying the above values with the line length, one can find the line total series impedance per
phase, and the line total shunt admittance per phase.
l z Z =
4-3
l y Y =
4-4
One can also define the transmission lines characteristic impedance and propagation constant
y
z
Z
c
=
4-5
y z =

4-6
As power systems consist of many lines, a simpler model of a power line is to describe each line by its -
equivalent, as shown in Figure 24:

Figure 24 - Equivalent -circuit of a transmission line

An analysis of this circuit shows that the -equivalent elements can be expressed as:
l
l
Z Z
L

) sinh(
=
4-7
2 /
) 2 / tanh(
l
l
Y Y
L


=
4-8
For a typical high voltage transmission line g can be neglected whilst r << x. By considering the lossless
line, i.e. neglecting r, the characteristic impedance is purely resistive, and the propagation constant is
purely imaginary:
C
L
Z
C
=
4-9
LC j =
4-10
With the previously mentioned assumptions, the reactive power loss in a transmission line can be
expressed according to equation 4-11:
2
L
Y
2
L
Y
L
Z
+
-
+
-
R
U
S
U
S
I
R
I
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 4
Transmission technologies

- 24 -

) sin(
) cos( ) cos( 2 ) cos(
2 2
l Z
l V V V l V
Q
C
R SR R S S

+
=
4-11
, where:
V
S
= Sending end voltage [V]
= The phase constant (= the value of the imaginary part of the propagation constant)
V
R
= Receiving end voltage [V].

SR
= Transmission angle, by which V
S
leads V
R
[]
Assuming V
S
V
R
V
N
, this equation can be rewritten as:
) )
sin
( 1 (cos
sin
2
) (
2
SIL
R SIL
R
P
l P
l
l
P
P Q


4-12
, where P
SIL
is the surge impedance load, defined as the power delivered at nominal voltage(V
n
) to a load
impedance equal to the characteristic impedance (Z
C
). It is given by:
C
n
SIL
Z
V
P
2
=
4-13
P
R
is the receiving end active power. For a more detailed derivation of these equations, see (Machowski,
Bialek, & Bumby, pp. 47-50). In Figure 25, Q is plotted as a function of P
R
/P
SIL
:

Figure 25 - Examples of reactive power absorbed by a lossless line as a function of its real load for various voltage ratings

When P
R
>P
SIL
, V
S
>V
R
, from Figure 25, it is clear that the overhead line is consuming reactive power. This
is the case when power is transmitted from node S to node R. The larger P
R
is, the larger the reactive
power is, which is in accordance with equation 4-12. The longer the line, the higher the value of the line
impedance, and thus the reactive power consumption is. To cope with these losses, one must feed reactive
power into the grid. This can be done for instance by connecting series or shunt capacitors to the line,
ideally so that the inductive part of the line impedance is equal to the capacitive part of the impedance.
A DC-transmission does not meet these problems. The reactance and susceptance of a line are
proportionally connected to the frequency of the system, and disappear for a DC-transmission. Even
though the conversion equipment is quite expensive, the economical benefit due to the non-existence of
reactive losses makes HVDC the preferred choice for overhead line distances longer than 600-800 km. It
can also be mentioned that HVDC does not suffer from skin effects as AC does. Also, for a given power
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 4
Transmission technologies

- 25 -

ratio, the peak value of the AC voltage is higher than the constant DC voltage. This means that for the
same conductor, more power can be transmitted through the same conductor area if it is DC instead of
AC.
4.1.2 Cables
A cable has a large capacitance per length unit. When the cable reaches a certain length, the value of the
capacitance is so large that the cables impedance can be considered purely capacitive. In such a case, the
cable only provides reactive power, due to the phase shift between the voltage and the current. The
possible length of the cable can be made longer with phase compensations in both ends of the line. Figure
26 shows the transmission capacity as a function of the transmission length for three different voltage
levels.

Figure 26 - Transmission capacity as a function of the transmission length for AC cables (Barberis, Todorovic, &
Ackermann, 2005)

By using HVDC for this purpose, no reactive power is produced or consumed in the cables. This means
that all of the cables transfer capacity can be used to transfer active power. For subsea cables, the length
where HVDC is more feasible than AC is approximately 50 km (ABB, 2008).
4.2 HVDC transmission
In an HVDC system, the electric power is converted from AC to DC in a converter station, transmitted to
the receiving end of the transmission, converted back to AC in a second converter station and injected into
the receiving AC system. It is used to connect areas which are not synchronous, like for instance the
Scandinavian system NORDEL and the western European system UCTE1. An HVDC transmission line
costs less than an AC line for the same transmission capacity. However, the terminal stations are more
expensive in the HVDC case due to the fact that they must perform the conversion from AC to DC and
vice versa. At a certain length of the transmission, called the break even distance, the HVDC solution
becomes more feasible than the AC solution. In addition to this, other factors like controllability and
reliability must also be considered when choosing the transmission technology.
The principle of an HVDC-transmission is shown in Figure 27. The three phase AC power is converted to
DC power in a rectifying circuit before it is transferred as DC current to the receiving end of the
transmission. At the end station, the DC power is inverted back to three phase AC power.
Thomas Haugsten Hansen
Offshore wind farm layouts


Figure
4.3 Traditional HVDC
A typical traditonal HVDC converter station
two poles, i.e. the link is bipolar. Each of the poles have the configuration as shown in
addition to the smoothing inductance
After passing the DC filter, the power is sent to the receiving end, shown in the f
terminal has the same configuration as terminal A, and the power is converted to AC. This AC power is
sent into power system B.

Figure 28 - The configuration of a typical bipolar twelve pulse HVDC conne
4.4 HVDC Light
HVDC Light, developed by ABB, is
technology uses semiconductors in the form of

Transmission technologies
- 26 -

Figure 27 - Principle of HVDC connection
onal HVDC converter station is shown in Figure 28. The converter station itself consists of
ch of the poles have the configuration as shown in Figure
addition to the smoothing inductance L
d
, DC filters help to minimize the ripple in the DC transmission.
After passing the DC filter, the power is sent to the receiving end, shown in the figure as terminal B. The
terminal has the same configuration as terminal A, and the power is converted to AC. This AC power is
The configuration of a typical bipolar twelve pulse HVDC connection(Mohan & Undeland, 2006)
, developed by ABB, is based on the use of Voltage Source Converters (VSC).
semiconductors in the form of insulated gate bipolar transistors (IGBTs)
Chapter 4
Transmission technologies
The converter station itself consists of
Figure 28. In
filters help to minimize the ripple in the DC transmission.
igure as terminal B. The
terminal has the same configuration as terminal A, and the power is converted to AC. This AC power is

(Mohan & Undeland, 2006)
based on the use of Voltage Source Converters (VSC). The
), and operate with
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 4
Transmission technologies

- 27 -

high frequency pulse width modulation in order to achieve high speed and, as a consequence, small filters
and independent control of both active and reactive power. It was introduced in 1997 and ranges from a
few tens of megawatts to the upper range of 1200 MW at 320 kV (ABB, 2008). ABB lists many
advantages with the technology; the electromagnetic fields are neutral, the cables are oil free and the
converter stations are more compact than for traditional HVDC. What is most relevant to this study is that
the technology increases the reliability of power grids.

Unlike conventional HVDC converters which usually require a 5% minimum current, the HVDC Light
converter can operate at zero power. The active and reactive powers are controlled independently, and at
zero active power the full range of reactive power can be utilized. Actually, at zero power, the HVDC
Light will have the properties of a Static Synchronous Compensator, or STATCOM (Machowski, Bialek,
& Bumby, 1998, p. 34). The active power transfer can be reversed very quickly with HVDC Light. The
reversion does not require converter blocking, filter switching or change of control mode, like traditional
HVDC. The power is reversed by changing the direction of the DC current instead of changing the
polarity of the DC voltage as in traditional HVDC technology.
The topology of an HVDC Light connection is the same as for traditional HVDC. Figure 27 shows this
principle. The areas in which HVDC Light disengages from traditional HVDC are the cable and the
converter stations. The HVDC Light cables use extruded polymer insulation while traditional HVDC
cables have used paper-oil insulated cables. This eliminates the risk of oil spillage. The cables are laid in
pairs with DC currents in opposite directions. This configuration eliminates the magnetic fields outside the
cables.

The complete topology of a converter station for HVDC Light is shown in Figure 29. The AC power is
transformed to the right voltage level and passes through an LC filter and a converter reactor before it is
converted to DC current in the converter valves. These are based on transistor technology, assuring very
good controllability of the active and reactive power. The DC current produced in the valves is filtered in
the DC capacitors before it is transmitted to the receiving end converter station. The DC current is then
inverted to AC, filtered in the converter reactor and LC filter, transformed to the voltage level of the grid
and sent into the receiving AC grid.

Figure 29 - HVDC Light configuration (ABB, 2008)
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 4
Transmission technologies

- 28 -

4.4.1 Converters
In the converter station, the AC power is rectified to DC power and vice versa. The IGBTs provide pulse
width modulation (PWM), which means that AC voltage can be produced by fast switching between
positive and negative DC voltage. Figure 31 shows this. Since the IGBTs are self commutating, the
amplitude and power angle of the AC voltage signal can be chosen freely. This means that the converter
can control the active and reactive power independent of each other. The six pulse transistor bridge is the
main part of the converter, as shown in Figure 30. On the DC side, capacitors are used to filter the
harmonics.

One of the disadvantages when it comes to this converter type is the high switching losses. For a
converter, the losses are proportional to the switching frequency according to the formula
sw loss conv
f k P =
;

4-14
In an HVDC Light converter, the switching frequency is 2 kHz, i.e. in one period in the 50 Hz system, the
voltage direction is reversed 40 times. This is shown in Figure 31. Traditional HVDC technology has line
commutated converters, which means that the switching frequency is the same as the line frequency. At
the pilot HVDC Light installation in 1997, the converter losses were approximately 4% of the total
capacity. Today, the losses have been reduced with 60%, and are now down to approximately 1.6% of the
total capacity for each converter. For traditional HVDC, these losses are 0.6-0.7% per converter.


Figure 30 - Six pulse transistor based converter configuration (Mohan & Undeland, 2006)


Figure 31 HVDC Light AC side voltage signal and the fundamental component of the harmonics (ABB, 2008)
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 4
Transmission technologies

- 29 -

4.4.2 Filters
An HVDC Light converter has a high switching frequency, as described in the previous chapter, and
shown in Figure 31. Because of the high switching frequency, there will only arise high frequency
harmonics. The shunt filters in Figure 29 are therefore relatively small. There is only need for a high pass
filter, unlike for the traditional HVDC solution, where tuned filters are also needed.
4.5 HVDC PLUS
HVDC PLUS (Power Link Universal System) is Siemens version of VSC HVDC technology. Just like
HVDC Light, HVDC PLUS is based on VSCs in the form of IGBTs. A consequence of the switching
technology used in the HVDC Light converters is that harmonics arise, resulting in high converter losses.
Siemens solution to this problem is to have a modular multilevel converter (MCC) consisting of several
modules. For each module, there is a capacitor. The topology is shown in Figure 32. This topology ensures
that each module get a small voltage level, without ripple. By controlling all the modules separately, the
output voltage is very close to a pure sine wave. The technology has less harmonics than HVDC Light,
and also a low level of high frequency noise. As a consequence of this, HVDC PLUS needs only small or
even no filters. This is a big advantage due to lower costs and lower maintenance requirement. In HVDC
PLUS converters, the switching frequency is only three times as high as the line frequency of the AC grid,
which gives smaller losses than for the HVDC Light switching.

As shown in Figure 32, the converter consists of three legs. Each leg is divided in series connected sub
modules. Each sub module consists of two IGBTs, a DC storage capacitor and an electronic control
system (illustrated by the gray arrows in the sub module shown in Figure 32). The topology provides the
possibility of using individual, selective control of the individual sub modules in a phase, and thus obtains
a voltage level selectable in small steps at the AC terminal, and, at the same time, a nearly constant DC
voltage at the DC terminals.


Figure 32 - HVDC PLUS output signal (left) and converter configuration (middle and right) (Siemens PTD, 2008)

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 5
Power system performance

- 30 -

5 Power system performance
In this master thesis, the term power system performance is used as a generic term for power system
stability, power system security and power system reliability. Even though these are related, they are not
the same. This chapter will give a presentation of each of the three terms and relate these to the operation
of an offshore wind farm. Details regarding the mathematical description of these phenomena are not
presented in this report. Instead, readers are referred to (Machowski, Bialek, & Bumby, 1998) and
(Kundur P. , 1994) for detailed descriptions of power system performance phenomena.
5.1 Power system stability
Power system stability is similar to the stability of any other dynamic system, and is based on the same
fundamental mathematics. It is simply stated an issue of remaining equilibrium between opposing forces.

The term denotes the ability of a power system to regain a state of operation equilibrium after being
subject to a physical disturbance, so that the system integrity is preserved. This means that practically the
entire power system shall remain intact, with no tripping of loads or generators, except for those
disconnected by isolation of the faulted elements or intentionally tripped to preserve the continuity of
operation of the rest of the system.

The disturbances might be of different nature, and they can be small or large. Small disturbances, such as
change in generation and load occur continuously. Large disturbances can for instance be the short circuit
of a line, or the loss of a large generating unit. A stable system will reach a new equilibrium after a
transient disturbance, while an unstable system will result in a run-away or run-down situation, for
instance a progressive decrease in system voltages or a progressive increase in rotor angle differences.
5.1.1 Classification of power system stability issues
The classification of power system stability issues facilitates the analysis of stability issues. By dividing
the issues into separate areas of interest, it is easier both to identify the factors that lead to instability and
to find effective methods to improve stable operation of power systems. Figure 33 shows how the IEEE
Power system dynamics committee classifies power system stability issues:
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 5
Power system performance

- 31 -


Figure 33 - Classification of power system stability. Based on(Kundur, et al., 2003)
As the figure shows, power system stability can be divided into three main topics; angle stability (or rotor
angle stability), frequency stability and voltage stability.
5.1.2 Rotor angle stability
Rotor angle stability denotes the ability of interconnected generators to operate in synchronism during all
operating conditions of the system. It depends on the ability of the machines in the system to maintain
equilibrium between the electromagnetic torque and mechanical torque of each generator in the system, so
if this kind of instability occurs, it is due to torque imbalance. This kind of instability occurs in the form of
increasing swings in the rotor angle of certain generators, finally leading to loss of synchronism with the
other generators in the system.

Small signal rotor angle stability issues are usually associated with the oscillations of one generator
against the rest of the system, called local plant mode oscillations. The stability of these issues depends on
the strength of the transmission system as seen from the generator, the generator excitation systems and
the plant output. Global problems might be caused by interactions among a large amount of generators in
the same area. These might oscillate as a group against another group of generators at another part of the
area. These oscillations are called interarea mode oscillations. The characteristics of such oscillations are
quite different from the local plant mode oscillations, and are very much affected by load characteristics.
The time frame for small signal rotor angle stability is in the order of 10 to 20 seconds following a
disturbance.
Power system
stability
Angle stability
Frequency
stability
Voltage stability
Short
term
Long
term
Small
signal
stability
Transient
stability
Short
term
Large
disturbance
voltage stability
Small
disturbance
voltage stability
Short
term
Long
term
- Ability to remain in operating condition
- Equilibrium between opposing forces
- Ability to maintain frequency within
nominal range
- Generation/load balance
- Ability to maintain synchronism
- Torque balance of synchronous
machines
- Ability to maintain steady voltages
-Reactive power balance
- Equilibrium of voltage control
Consideration
for classification
Physical nature/
main system
parameter
Size of
disturbance
Time span
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 5
Power system performance

- 32 -


Transient stability issues are associated with the ability of the power system to withstand a large
disturbance, such as line tripping or the short circuit of a line. When such a fault occurs, large currents and
torques are produced, and often action must be taken quickly to remain system stability. The instability is
usually related to insufficient synchronizing torque, resulting in first swing instability. However, in large
power systems, transient instability may also be the result of nonlinear effects affecting a single mode
causing instability in a preceding swing. It could also be a result of superposition of a slow interarea swing
mode and a faster local area swing mode, resulting in rotor angle instability.
The time frame for transient stability studies is usually up to five seconds following the disturbance, but
for very large systems, the time span of interest may extend to as much as 20 seconds.
5.1.3 Frequency stability
Frequency stability denotes the ability of the system to maintain nominal frequency following a
disturbance of the system leading to an imbalance between load and generation. Frequency instability
occurs in the form of sustained frequency swings that lead to the tripping of generators and/or loads. In a
small system, frequency instability could be of concern for any large disturbance causing a significant loss
of generation or load, while in a large system, this kind of problem is only of concern if a severe
disturbance occurs, dividing the system into separate islands operating independently of each other. In
islanded operation, frequency instability may occur for any disturbance causing a relatively large loss of
load compared to the total island load.
Frequency stability problems are associated with inadequacies in equipment responses and poor
coordination of control and protection systems. During frequency mismatches, the characteristic time of
the process may vary from hundreds of milliseconds, corresponding to generator excitation systems,
protection or load shedding, up to minutes, corresponding to the response of devices like for instance load
voltage regulators or prime movers. Thus, frequency stability issues can be divided into short term or long
term issues.

Short term frequency stability issues can be the formation of an undergenerated island with insufficient
underfrequency load shedding, causing a blackout of the system within seconds. The time of interest is
from hundreds of milliseconds up to seconds.
Long term frequency stability issues may be a more complex situation, where for instance the control
system of a power plant is incorrectly tuned so that there is inadequate overspeed controls. The time of
interest is from tens of seconds up to minutes.
5.1.4 Voltage stability
Voltage stability denotes the ability of the system to maintain steady voltages at all buses following a
disturbance from a given initial condition. Voltage instability is the progressive and uncontrollable rise or
fall in voltage of one or more buses due to a disturbance from the initial operating conditions. The main
factor causing voltage instability is the inability of power systems to maintain a proper balance of reactive
power and voltage control actions. The driving force for voltage instability is usually the loads. Following
a condition of reduced transmission system voltages, the power consumed by the loads tends to be
restored by the action of thermostats, tap changing transformers or voltage regulators. Voltage instability
may lead to loss of load or loss of system integrity.

Short term voltage stability involves the dynamics of fast acting power system components, such as
induction generators, power electronic devices and electrically controlled loads. The time interval for short
term voltage stability issues is in the seconds range. This type of analysis is based on sets of differential
equations, and the dynamic modeling of loads is therefore often essential to get good results.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 5
Power system performance

- 33 -

Long term voltage stability involves slower working elements, such as tap-changers, thermostatically
controlled loads and generator current limiters. The time interval of interest may extend to several
minutes. The long term voltage stability of a system is often determined by the outage of equipment,
rather than the severity of the initial disturbance. Instabilities occur due to long term imbalance, for
instance when loads try to restore their power beyond the capability of the transmission network feeding
the load, the post-disturbance steady state operating point being small-signal unstable, or a failure to reach
the post-disturbance steady state operating point. Also, long term load buildups can cause this kind of
instability.
5.2 Power system security
Power system security denotes the ability of a power system to survive disturbances without the
interruption of customer service. To be secure, the system must, in addition to being stable, be secure
against contingencies that are not classified as stability problems, for instance sabotage, fall of
transmission towers due to ice loading or an explosive failure of a cable. Also, security includes the issue
of the consequences of instability. Two systems might both be stable and have the same stability margins,
but unequally secure due to the fact that the consequences of instability are more severe in one system
than the other.

The security analysis of a power system is made to determine the robustness of the system when subject to
one or more of the already mentioned disturbances. There are two important components of security
analysis; Static security analysis and dynamic security analysis. Static security analysis involves steady
state analysis of post disturbance system conditions to determine whether the equipment ratings or voltage
constraints are violated. Dynamic security analysis involves examining different categories of stability
issues as described in chapter 5.1. Hence, stability analysis is an integrated part of security analysis.

A common way to perform a security assessment is to use the N-1-criterion. For an N-component power
system to fulfill the N-1-criterion, the system must withstand the loss of any of its components and still be
in a stable steady state operation mode. In practice, this means that the disconnection of the largest power
plant should not cause the disconnection of any consumer. The criterion can be checked by defining a
certain amount of contingencies that have a significant likelihood of occurring, and study how these will
affect the system by simulating it in a power system simulation tool.
5.3 Power system reliability
Power system reliability deals with the probability of satisfactory operation over the long run. It denotes
the ability to supply adequate electrical service on a nearly continuous basis, with few interruptions over
an extended time period. To be reliable, the system must be secure most of the time. While security and
stability are time-varying issues that can be judged by studying the performance of the power system
under a particular set of conditions, for instance by using simulation tools, there is a need for probability
distributions and consequence analysis when performing a reliability analysis.

Reliability can be addressed by studying two basic functions of the power system; Adequacy and Security.
Adequacy is the ability of the power system to supply the electric power demanded by the customer at all
times, accounting for possible outages of system components. Security is already described.

In order to estimate reliability indices, one must be able to predict the system behavior. The components in
the system, such as cables, transformers, generators and gearboxes, can either be functioning (100%
operability) or out of function (0% operability). The mean time to failure (MTTF) is the mean time spent
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 5
Power system performance

- 34 -

in the on-state, i.e. the time the component is working between two faults. The mean time to repair
(MTTR) is the mean time it takes to repair a fault on the component in question or replace the faulted part,
i.e. the mean downtime when a fault occurs. The mean time between failures (MTBF) is the sum of the
mean time to failure and the mean time to repair. This is illustrated in Figure 34:


Figure 34 - The failure and reparation process of a component

When it comes to offshore wind power plants, reliability is even more important than for onshore wind
power plants. The consequences of a fault offshore may be severe, with reparation times up to several
months and planned power ratings up to 1000 MW. Up until today, providing redundancy has not been
considered for existing wind farms since the likelihood for a fault and the associated costs has been
assumed to be lower than the costs of the extra equipment needed to provide redundancy. As wind farm
sizes increase, the amount of energy (and income) lost might be high enough for redundancy to become
profitable. Cost/benefit calculations regarding redundancy should be performed for all projects.

In a Garrad Hassan study (Gardner, Craig, & Smith), based on statistical data from 1950 to 1980, a cable
failure rate of 0.32 failures/100km of cable/year was found. Since these data are old, the study suggests a
cable failure rate of 0.1 failures/100km of cable/year instead. In the literature study conducted as a part of
this thesis, it was found that estimates for cable failure rates vary from 0.08 to 0.32 failures/100km/year.

Another important reliability issue is related to the capacity margin of the power system. There must be
enough capacity available in the power system to cover the peak load. Adding wind power to the power
system the installed power increases, meaning that the capacity of the producers in the grid also increases.
The capacity factor of a wind farm (CF) depends on the wind resources and the type of wind turbine. It
denotes the average power as a percentage of the nominal capacity, and can be expressed as:

% 100
8760

=
m
P
W
CF
5-1

, where W is the annual energy production, and P
m
is the mean power production of the wind farm, which
can be expressed as:

=
IC
C
p m
dx x f x P
0
) (
5-2
MTBF
MTTF MTTR
100%
0%
Time
Operability
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 5
Power system performance

- 35 -

C
IC
= total installed capacity [MW]
x = Production level of the wind farm [MW]
f
p
(x) = probability density function of the total wind power from the wind farm over one year

The CF of most wind power plants on land lies between 20% and 40%, which can be expressed as 1800
3500 h/a.. Offshore wind power plants can reach up to 4000-5000 full load hours, corresponding to a CF
of 45% - 60%. In comparison, combined heat and power has full load hours in the range 4000-5000 h/a,
coal power plants has 5000-6000 h/a. and nuclear power plants have 7000-8000 h/a (Ackermann, 2005, p.
149). This shows that moving wind power plants offshore might improve the reliability of the wind power
production.
5.4 Grid codes
There are no specific grid codes regulating the operation of the internal grid of an offshore wind farm
directly. Nevertheless, all countries have specified grid codes, regulating the operation of the national
electricity grid, and the offshore wind farm will have to fulfill the operational criterions described in these
codes. The grid codes are decided by the TSO, and might vary from country to country The Norwegian
TSO, Statnett, has defined a set of grid codes (Statnett SF, 2008) that must be fulfilled by all power plants
in the Norwegian power system, and in this thesis the Norwegian grid codes are used as reference. In the
following it is described how these codes apply for the offshore wind farm, relating them to the power
system performance issues described in the previous chapters.

For an offshore wind farm, rotor angle stability is a matter of maintaining synchronism for all wind
turbines. The fault ride through demands for a power plant coupled to a point in the Norwegian power
system with voltage above 200kV state that the power plant needs to withstand a voltage drop to 0 pu
lasting up to 150 ms (Figure 35). For a fault occurring in the offshore grid, there is no such demand. The
main idea of improving transient stability of a generating unit is to decrease the available energy for
acceleration of the generating unit during a grid disturbance.

As equations 2-7 to 2-14, show, the rotor angle is highly dependent on the power output of the wind
turbine. Hence, rotor angle control can be done by rapidly changing the output power of the generators in
the system. Equation 2-11 shows that to control the electrical output power P
e
one can lower the output
voltage by applying voltage control, or one can control the stator currents. The predefined generic wind
model in PSS/E

contains control systems for the DFIG. These are described in Appendix 4.

When it comes to frequency stability, this is a matter of keeping the frequency in the PCC within the
descriptions given in the grid code. The permitted stationary frequency deviation in the Nordic system is
0.2% = 0.1 Hz. The permitted frequency drop during a dimensioning fault is 1%, or 0.5 Hz, i.e. the
frequency is not allowed to drop below 49.5 Hz. In the Nordic system this control is above all situated in
the hydro generators. Grid frequency oscillations can be counteracted with controlled power injection into
the grid. For the wind farm to be able to provide this, the need for active power regulation of the wind
farm arises, either in the turbines or by controlling the active power transmission in the transmission
system. It must be possible to adjust the production to any value in the area 20-100% of nominal power,
and it must also be possible to regulate the production from nominal output power to stop in maximum 30
seconds (Statnett SF, 2008). As a general comment, it should be noted that in the Norwegian power
system, the use of wind power plants as frequency regulators would not be a good solution as long as there
is hydro power available. Since there is no storage opportunity for wind power, the alternative value for
the wind is zero. The wind power production should therefore be kept at maximum at all times. The
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 5
Power system performance

- 36 -

frequency regulation should therefore be performed by hydro power plants as long as that is an
opportunity. Nevertheless, in islanded operation, active power control of the wind farm can be necessary.
Also, as the amount of wind power in the grid increases, the offshore wind farms must also contribute to
frequency stabilization, not just be able to stay connected to the grid.

Table 3 shows the boundaries for the operational area of power plants, regarding frequency, voltage level
and time duration. Within these boundaries, the power production plant must be able to operate
continuously for the time requirements stated in the table.

Frequency [Hz] Voltage [pu] Time
45.0 47.5 0.90 1.05 >20 s
47.5 49.0 0.90 1.05 >30 min
49.0 52.0 0.90 1.05 Continuously
52.0 53.0 0.90 1.05 > 30 min
53.0 55.0 0.90 1.05 >20 s
55.0 57.0 0.90 1.05 >10 s
Table 3- Functional area of power production plants (Statnett SF, 2008).
Operating an induction machine at a significantly different frequency than what it is designed for can
cause a significant change in the magnetizing reactance of the machine and, because of this, an out-of-
proportion increase in magnetizing current. The net result might be overheating of the machine windings.
To prevent this, the ratio of volts per hertz in the machine should be kept constant, or expressed
mathematically:
fnom
nom
f
V
f
f
V
2
2
=
5-3
fnom
nom
f
P
f
f
P
2
2
=
5-4

, where f
nom
denotes the nominal frequency (50 Hz) and f
2
denotes the frequency when there is a frequency
deviation in the system.

The main voltage stability issue for an offshore wind farm is the fault ride through demand. For
connection points where the voltage is higher than 200 kV, the Norwegian TSO, Statnett, has the
following fault ride through demands to power plants connected to the grid: The power plant has to stay
connected and deliver power at a voltage at the connection point above the voltage profile described as
follows:

Voltage reduction to 0 pu for up to 150 ms, followed by instant voltage increase to 0.25 pu.
Linear voltage increase from 0.25 pu to 0.90 pu for 750 ms
Constant voltage at a minimum value of 0.90 pu

The offshore wind farm will have to supply a voltage that meets these demands, which are shown
graphically in Figure 35 :
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 5
Power system performance

- 37 -


Figure 35 - Fault ride through requirements for power plants connected to a grid point with voltage >200kV in the
Norwegian grid (Statnett SF, 2008)
In addition to meeting the fault ride through demands, the wind farm is required to keep the voltage level
in the range of 0.9 pu to 1.1 pu at the connection point at all other times.

Another issue when it comes to voltage stability is the control of reactive power. According to the grid
codes, the wind farm at nominal production shall have a reactive power rating at a level where the power
factor is kept between 0.95 leading and 0.95 lagging. At production below rated power, there shall be no
limitations when it comes to using the reactive capacity of the plant. Figure 36 shows the reactive power
regulation area for wind power plants:

Figure 36 - Reactive power capacity for wind power plants in the Norwegian grid (Statnett SF, 2008)
With an HVDC transmission to shore, this demand is taken care of by the onshore converter station
control. With an AC transmission system, this must be taken care of by adjusting the power factor of the
wind turbines, or by applying some kind of reactive power compensation device in the PCC.


Time after fault (s)
V
o
l
t
a
g
e

i
n

c
o
n
n
e
c
t
i
o
n

p
o
i
n
t

(
p
u
)
Inductive Power factor Capacitive
P/Pn
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 6
Simulation models

- 38 -

6 Simulation models
This chapter describes the setup of the simulations that are executed in this thesis. A short presentation of
the simulation tool that is used to perform the simulations is given, and the choice of parameters and
models are presented and justified.

All simulations have been executed in the program PSS/E (Power System Simulator for Engineering).The
program, which is developed by Siemens PTI (Power Technologies International), is widely used in
electrical transmission planning for static and dynamic analysis of power systems. In this thesis it is used
to perform load flow calculations and to study the dynamic behavior of the wind farm layouts that are
being compared.
6.1 Physical parameters for the wind farm site
The wind farm is placed in the North Sea and connected to the Norwegian grid. It is assumed that the wind
turbines must be placed at water depths shallow enough for existing installation technologies to be used,
i.e. floating offshore wind turbines are not considered. For the existing technologies to be utilized, the
wind farm should be placed outside the Norwegian trench. The trench is from 50 to 95 kilometers wide,
and therefore a transmission distance of 100 km is chosen in this thesis. This distance is from the onshore
PCC to the offshore substation.

The wind farm consists of 108 wind turbines rated at 5MW. Most sources suggest that the distance
between the turbines should be three to four rotor diameters in the direction perpendicular to the
dominating wind direction, and five to nine rotor diameters in the dominating wind direction, to minimize
wake losses. Using data from an existing 5 MW wind turbine, the RE power 5M wind turbine (RePower
Systems AG), a rotor diameter of 126 m is used for the turbines. Based on this diameter, the distances
between the wind turbines are chosen to be 500 m in the direction perpendicular to the dominating wind
direction and 900 m in the dominating wind direction. The electrical connection of the wind turbines is
done by dividing the wind farm into twelve feeders consisting of nine turbines each, where all turbines in
one feeder are connected to each other either in a radial, looped or star design, as described in chapter 3.5.
It is assumed that the distance from the first wind turbine in the feeder to the offshore substation is 5 km.
6.2 Static models
The cables, transformers, generators, shunt elements and loads are modeled using the predefined models
in PSS/E . For all simulation cases, there is only one load. This is placed at the onshore swing bus, which
is representing the onshore power system. For the systems using AC transmission to shore, the SVC and
fixed reactors are designed so that zero reactive power is delivered to the PCC.
6.2.1 Wind turbines
Version 31 of PSS/E includes a wind turbine model, which is used for all wind turbines studied in this
thesis. The data for the wind turbines are the same for all simulations. All wind turbines are rated at 5
MW, with a maximum power factor of 0.95. This gives an apparent power-rating of 5.26 MVA for the
turbines. PSS/E version 31 includes three options for reactive power control of the turbines:

1) The wind machine participates in voltage control, with user defined reactive power limits. The
values of QT and QB on the data record in PSS/E specify the machines reactive power limits.

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 6
Simulation models

- 39 -

2) The wind machine participates in voltage control, with a specified power factor and the machines
active power setting (PG on the data record in PSS/E ) used to set the machines reactive power
limits.

3) The wind machine operates at fixed active and reactive power. The machines reactive power
output and reactive power upper and lower limits all equal, and set based on the specified power
factor and the machines active power setting (PG on the data record in PSS/E ).

Option 2) is chosen for all simulations in this thesis.
6.2.2 Transformers
All transformers in the system models have impedance values based on the transformers MVA rating.
(Hubert, 2002, p. 72) determines that the efficiency of large transformers may be higher than 99%, and
based on this, all transformers are designed to have active power losses in the range of 0.5-1.0%. The
parameter values for the transformer resistance and inductance for all transformers
1
are:

R = 0.007 pu
X = 0.1 pu
6.2.3 Cables
To decide the cable parameters of all cables in the system models, two parameters are considered; the
cable length and the power rating of the cable.

For all models, the cable length is given by the physical parameters of the wind farm. The distances
between the wind turbines are 500 m in the direction perpendicular to the dominating wind direction and
900 m in the dominating wind direction. The distance from the first wind turbine in the feeder to the
offshore substation is 5 km. The cable to shore is 100 km.

When it comes to the power rating of the cable, this is decided by looking at the maximum power it must
be able to handle in a worst case scenario. It is assumed that no tapering of cable capacity is done, and
thus all cables in one feeder must be able to carry the worst case scenario power. The offshore cable
ratings differ from layout to layout. Table 4 shows the cable parameter values for the AC cables used in
this thesis:

Cable type Manufacturer U
nom
I
nom
r [(km] x [(km] c [F/km]
1x3x95 XLPE-M-AL-LRT Ericsson 11 265 0,32 0,11 0,34
TSLE 3x1x800 Al/- Nexans 33 780 0,037 0,16 0,36
TSLE 3x1x2000 Al/- Nexans 33 1050 0,015 0,14 0,53
TSLE 3x1x150AQ Nexans 48 1025 0,0186 0,09 0,13
XLPE 3x1x 1000 CU-LRT Ericsson 400 1220 0,0151 0,12 0,17
Table 4 - Cable parameters of the AC cables used in the different wind farm layouts


1
The transformers that are included in the ABB HVDC Light-model have parameter values as stated in the classified
information from ABB. These values are not recited in this thesis.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 6
Simulation models

- 40 -

6.2.4 HVDC Light
The user model of HVDC Light is developed and provided by ABB. The standard elements from PSS/E
are used in order to establish the load flow. Table 5 shows the available modules for HVDC Light
technology: All parameters are taken from the ABB document Its time to connect (ABB, 2008).

Table 5 - HVDC Light modules(ABB, 2008)
For all load flows, the module M6 is used to represent the HVDC Light transmission. This has a rated
power, S
base
, of 570 MVA.

The rectifier is modeled as a PU bus connected to a generator and shunt. Because the power is going from
the AC to the DC side, the specified power to the generator has to be negative. The generator impedance
Z
source
in PSS/E specifies the converter reactance in Figure 29. The shunt AC filter in the same figure is
represented by the reactive power generation of the fixed shunt capacitor at the rectifier bus in PSS/E .
Additionally, a converter transformer has to be added. One side is connected to the rectifier bus, and the
other side is connected to the offshore substation bus. The AC voltage at the converter bus is 195 kV for
the HVDC Light converter in module M6. The inverter is modeled using the same units as the rectifier,
but the power from the generator has to be positive, because the power is going from the DC side to the
AC side.

The reactive power limits to be specified in the power flow for the generator equivalents depend on the
active power. The HVDC Light converter has a capability curve according to Figure 37:.

Figure 37 Capability curve for HVDC converter
It is up to the user to select the right reactive power limits for the converter, based on this curve.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 6
Simulation models

- 41 -

The DC cable is not modeled in the load flow. The power loss in the HVDC Light system must be
calculated, and the received power at the onshore converter bus generator must be specified based on this
calculation in order to represent the HVDC losses correctly in the load flow simulation. Equation 6-1
describes the power loss in the HVDC Light system.

inverter rectifier DC loss
P P P =
,

6-1

The power loss in the DC system can be split into converter losses and line losses. Table 6 shows the full
power ratings of HVDC Light modules M4 to M6:


Table 6 Sending and receiving power in the 150kV HVDC Light modules
From Table 6, the converter losses at full power can be calculated by comparing the sending power with
the receiving power when the system is has a back-to-back configuration.

The dynamic converter losses are calculated as a constant part, no load losses, and a load loss which is
estimated to be linear with the load

The no load losses are estimated as 0.3 pu of the nominal losses
The load losses are estimated as 0.7 pu at nominal load (Sbase)
The nominal losses are estimated to 0.0165 pu of Sbase, according to (ABB, 2008)

This can be expressed mathematically by equations 6-2 to 6-4:
base nom
S P = 0165 . 0
6-2
nom noload
P P = 30 . 0
6-3
max ,
7 . 0
sent
sent
nom load
P
P
P P =
6-4
Total converter losses for one converter can be expressed as:
noload load conv loss
P P P + =
,

6-5
For a bipolar transmission, the DC cable loss for can be calculated by finding the DC current, given a
certain cable dimension and length. The DC current in each cable can be expressed as:
base
conv loss sent
DC
U
P P
I

=
2
1000
,

6-6

In (ABB, 2008), an overview of the resistance for different cables for different conductor areas is
presented. From Table 6, it can be found that the copper conductor area of the DC cable needs to be 2800
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 6
Simulation models

- 42 -

mm
2
. Looking at the resistance overview, the resistance is found to be 0.0079 /km. With a 100 km
distance to shore, the total resistance of the cable is 0.79 This can be used to calculate the cable losses
as:

2
,
2
DC cables loss
RI P =
6-7

Finally, the received power at the inverter bus can be expressed as:

cables loss conv loss sent received
P P P P
, ,
2 =
6-8
6.2.5 Onshore generator and SVC
The onshore power system is modeled as one bus with one generator and one load. The generator
representing the power system is modeled by the predefined generator model in PSS/E . In addition to the
generator at the PCC, a load of 540 MW is connected at the same bus. This way, the active power
production at the generator in the PCC equals the total active power losses of the wind farm.

For the AC/AC-models, an SVC is placed at the onshore transmission bus. The SVC is modeled using the
standard generator model, defining the active power output as zero.
6.2.6 Fixed shunts
For the AC/AC-models, two fixed shunt reactors are placed at the onshore transmission bus to provide
reactive power compensation to the AC cables. The shunts are modeled using the predefined fixed
shunt-model in PSS/E .
6.3 Dynamic models
All generators and power electronic devices in a power system consist not only of the device itself. To
make sure that the system fulfills the system requirements, all devices are equipped with control systems
such as automatic voltage regulators (AVRs), power factor correctors (PFCs) and power system stabilizers
(PSSes). Thus, for PSS/E to run dynamic simulation, the program demands that dynamic models are
defined for all generators, SVCs and VSC HVDC lines. In the following, the models used in this thesis are
presented.
6.3.1 Wind turbine model
In this thesis, the DFIG is chosen to represent the wind turbine generators. PSS/E contains a predefined
model of this turbine type, the WT3 generic wind model, which comprises the following modules:

WT3G: generator/converter module
WT3E: electrical control module
WT3T: mechanical control (wind turbine) module
WT3P: pitch control module.

The model was developed to simulate performance of a wind turbine employing a doubly fed induction
generator (DFIG) with the active control by a power converter connected to the rotor terminals. The model
is described in detail in the application guide of PSS/E (Siemens PTI, 2007), and the model is described
and all block diagrams reproduced in Appendix 4. The chosen parameter values for the wind turbine
model are also given in the appendix. When conducting the DC simulations, the control parameter
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 6
Simulation models

- 43 -

VLTFLG (see the block diagram of the WT3E1 module in appendix 4) had to be changed as compared to
the AC simulations. VLTFLG had to be equal to 1 for the AC/DC simulations, and 0 for the AC/AC
simulations, in order to get the initial conditions check to be OK.
6.3.2 Onshore generator model
The onshore generator model is modeled as a salient pole hydro generator, since the Norwegian power
system is based on almost only hydro power generation. To represent the dynamic behavior of the
generator, the model GENSAL is used. The exciter and governor systems are represented by the models
SCRX and HYGOV, respectively. All parameters can be found in appendix 4.
6.3.3 SVC model
The SVC used in the simulations is modeled as a generator, producing zero active power during the load
flow simulations. In the dynamic simulations, the PSS/E model CSVGN5 is used. All parameters can be
found in appendix 4.
6.3.4 HVDC Light model
In order to represent the dynamic behavior of the HVDC Light system, ABB has developed
two user models called CABBL2 and CEMPTY. Each module presented in Table 5 has its own dynamic
file, which is given in the user manual provided by ABB(ABB Power technologies AB, 2009). The
dynamic file corresponding to the module M6 is applied for the dynamic simulations in this thesis.

Figure 38 shows the structure of the load flow and dynamic model of HVDC Light.

Figure 38 - The HVDC Light model used in PSS/E

In the load flow, the PSS/E generic generator model represents the converters. The user model CABBL2
is used to calculate the current injection to be applied by this generic generator. This user model includes a
representation of the cable, including the cable resistance. The user model CABBL2 is applied as the
primary dynamic model for the generator used as the first converter in PSS/E. Still CABBL2 controls
both the first and the second converter. An additional dummy user model CEMPTY is applied as the
primary dynamic model for the generator used as the second converter. Without this dummy user
model, PSS/E will disconnect the second converter in the dynamic simulation.
.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 44 -

7 Load flow simulations
This chapter describes the load flow simulations executed as a part of the work with this thesis. For each
wind farm layout, two cases are studied:

In the first case, the simulation is setup so that the active power delivered to the PCC is maximized. This
is done by varying the voltage level at the offshore transmission bus and choosing the voltage level that
gives the highest amount of active power delivered to the PCC, given that all elements of the system is
rated below 100%, and that the voltage level at all buses does not exceed its upper and lower limit. For
each simulation, PU-curves, showing the optimum voltage/power-level have been created. The PU-curves
are given in appendix 2. This load flow situation is from now on referred to as normal operating
conditions.

In the second case, it is assumed that a fault has occurred at the most critical cable in one of the feeders.
This fault has led to the cable being disconnected. The voltage at the offshore transmission bus is the same
as before the fault occurred. This load flow condition is from now on referred to as worst case operating
conditions.

In chapter 3.5, six options for the design of the offshore grid were presented, and these are all studied by
running load flow simulations on them. In addition, the impact of the choice of transmission system to
shore is studied. Two options are studied. The first option is to use AC transmission, using two AC cables
to provide redundancy. The second option is to use HVDC Light transmission, using two HVCD links to
provide redundancy. Having six offshore grid designs, two transmission system options and two load flow
cases, this means that a total of 24 load flow simulations are executed in this thesis.
7.1 Large AC wind farm layouts
The large AC wind farm layout was presented in chapter 3.2.2 . It consists of an offshore AC network
with an AC transmission to shore. Figure 39 shows the transmission system to shore. A 33/400 kV
transformer provides the right voltage level. The cables are 400 kV XLPE cables, rated at 845.2 MVA.
The cable data are gathered from an Ericsson datasheet, available at their web page (Ericsson nkt).

Figure 39 - Single line diagram showing the connection to shore, using two AC cables to provide redundancy
At the onshore side, there is a considerable need of reactive compensation. This is taken care of by having
a constellation consisting of two fixed shunt reactors and one SVC at the onshore transmission bus. The
constellation is designed so that the SVC provides zero reactive power at unity power factor at the PCC.
The SVC assures that the reactive capacity requirement of being able to provide cos = 0.95 at nominal
power output is fulfilled. At normal operating conditions, the offshore wind farm delivers 515 530 MW
to the PCC, depending on the wind farm layout. With a power factor limit of 0.95, the SVC must be rated
at a minimum of 170 175 MVAr. The total reactive power delivered to the onshore transmission bus
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 45 -

from the two AC cables is in the range of 1611 to 1613 MVAr for the simulation cases. The reason why
this is so high is the high capacitive rating of the cable (0.20 F/km, giving 17 F in total per cable),
combined with the fact that there is no reactive compensation offshore.

The fixed shunts are rated at the same reactive power level as what is delivered to the onshore
transmission bus by the AC cables. If one of the transmission lines is lost, one of the fixed shunts are
disconnected, so that the SVC still does not have to provide reactive power compensation at unity power
factor. This running condition is indicated in Figure 40 :

Figure 40 - Single line diagram showing the connection to shore with one AC cable disconnected
For all simulations, the disconnected shunt varies between 845MVAr and 846 MVAr, while the shunt still
in connection varies between 765 and 766 MVAr. This situation is not studied as a load flow case, only
during the dynamic simulations.
7.2 AC/DC wind farm layouts
The AC/DC wind farm layout was presented in chapter 3.3. In the simulations, the AC/DC configuration
consists of an offshore AC network with a double HVDC Light transmission to shore. Traditional HVDC-
technology is not studied. Figure 41 shows the transmission system to shore.

Figure 41 - Single line diagram showing the connection to shore, using two HVDC Light connections
A 33/132 kV transformer offshore boosts the voltage before the power is converted from AC to DC in the
HVDC Light converter station. Onshore, the power is transformed from 195 kV DC voltage to 300 kV AC
voltage before it is transmitted to the PCC. The DC cables are 195 kV XLPE cables, rated at 570 MVA.
7.3 Radial design
In the radial design, nine wind turbines are connected in a radial, as described previously. The wind
turbines are rated at 5MW, with a power factor limit of 0.95. The wind turbines are operating at 0.69kV.
Each wind turbine is equipped with a 0.69/33kV transformer, rated at 5.26 MVA. The cable ratings are
48.6 MVA throughout the loop, operating at 33 kV. The maximum rated voltage of the cable is 36 kV, or
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 46 -

1.091 pu. Thus, all voltage levels lower than 1.091 pu are considered acceptable in the offshore grid. A
value of 0.9 pu is used as the minimum acceptable voltage level.
7.3.1 AC/AC
Figure 42 shows the load flow result during normal operating conditions for one feeder of the radial
design while using AC transmission to shore. All elements of the feeder operate at below their maximum
rating. The most heavily loaded cable is the one connecting the first wind turbine to the offshore
substation. This is loaded at 92%. The most heavily loaded devices in the feeder are the wind turbines,
which operate at 95% of their maximum rating. All voltage levels are within their boundaries.


Figure 42 - Radial design with AC transmission normal operating conditions
The active power delivered to the grid is 524.6 MW, giving total losses of 2.85%. The most heavily
loaded elements of the system are the transmission cables, rated at 100%.

Figure 43 shows the load flow result during worst case operating conditions for one feeder of the radial
design while using AC transmission to shore. All elements of the feeder operate at ratings below their
maximum rating. The most heavily loaded devices in the feeder are the wind turbines and wind turbine
transformers, which operate at 95% of their maximum rating.


Figure 43 - Radial design with AC transmission - worst case operating conditions
During worst case operating conditions, that is, during operating conditions where one radial is out of
service, the active power delivered to the grid is 480.9 MW. This means that the total losses are 10.94%
when compared to the total installed power of the wind farm, 540 MW. The SVC must consume 6.2
MVAr in order to provide unity power factor in the PCC.
7.3.2 AC/DC
Figure 44 shows the load flow result during normal operating conditions for one feeder of the radial
design while using AC transmission to shore. All elements of the feeder operate at below their maximum
rating, but slightly higher than in the AC/AC-system. The most heavily loaded cable is the one connecting
the first wind turbine to the offshore substation. This is loaded at 94%. The most heavily loaded devices in
the feeder are the wind turbines, which operate at 98% of their maximum rating. All voltage levels are
within their boundaries.


Figure 44 - Radial design with DC transmission normal operating conditions
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 47 -

The active power delivered to the grid is 502.7 MW, giving total losses of 6.91%. The most heavily
loaded elements of the system are the wind turbines, at 98%. The generators representing the HVDC Light
connection are running at 48% and 46% at the offshore and onshore stations respectively (shown in
appendix 2).

Figure 45 shows the load flow result during worst case operating conditions for one feeder of the radial
design while using AC transmission to shore. All elements of the feeder operate at ratings below their
maximum rating. The cable ratings are as in the normal operating conditions. The most heavily loaded
devices in the feeder are the wind turbines and the wind turbine transformers, which operate at 97% of
their maximum rating. All voltage levels are within their boundaries.


Figure 45 - Radial design with DC transmission - worst case operating conditions
During worst case operating conditions, which is during operating conditions where one feeder is out of
service, the active power delivered to the grid is 460.2 MW, corresponding to 14.78% losses when
comparing to the nominal power output. The most heavily loaded elements of the system are the wind
turbines and the wind turbine transformers. The generators representing the HVDC Light connection are
running at 45% and 43% at the offshore and onshore stations respectively (shown in appendix 2).
7.4 Single sided ring
In the single sided design, nine wind turbines are connected in series, with a safety cable from the last
wind turbine to the offshore substation. The wind turbines are rated at 5MW, with a power factor limit of
0.95. The wind turbines are operating at 0.69kV. Each wind turbine is equipped with a 0.69/33kV
transformer, rated at 5.26 MVA. The cable ratings are 48.6 MVA throughout the loop, operating at 33 kV.
The maximum rated voltage of the cable is 36 kV, or 1.091 pu. Thus, all voltage levels lower than 1.091
pu are considered acceptable in the offshore grid. A value of 0.9 pu is used as the minimum acceptable
voltage level.
7.4.1 AC/AC
Figure 46 shows the single sided ring design during normal operating conditions. Due to the extra cable
connection from the last wind turbine in the feeder to the offshore substation, none of the cables are rated
higher than 55%. The most heavily loaded devices are the wind turbine transformers, which are rated at
96%. All voltage levels are within their boundaries.


Figure 46 - Single sided ring, normal operating conditions
The active power delivered to the grid is 526.6 MW, giving total losses of 2.48%. The most heavily
loaded elements of the system are the transmission cables, rated at 100%.

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 48 -

Figure 47 shows the worst case load flow situation for the single sided ring design. The line between the
first wind turbine and the offshore substation is lost, meaning that the direction of the power flow is
reversed through the feeder. The most heavily loaded cable is now running at 91% of its current rating.
The most heavily loaded devices of the feeders are still the wind turbine transformers, at 96%. All voltage
levels are within their boundaries.


Figure 47 - Single sided ring, worst case operating conditions
The active power delivered to the grid is 526.2 MW, corresponding to 2.56% losses when comparing to
the nominal power output. The most heavily loaded elements of the system are the transmission cables,
rated at 100%. The SVC must consume 0.1 MVAr in order to provide unity power factor in the PCC.
7.4.2 AC/DC
Figure 48 shows the single sided ring design during normal operating conditions. Due to the extra cable
connection from the last wind turbine in the feeder to the offshore substation, none of the cables are rated
higher than 55%. The most heavily loaded devices are the wind turbine transformers, which are rated at
96%. All voltage levels are within their boundaries.


Figure 48 - Single sided ring, normal operating conditions
The active power delivered to the grid is 504.6 MW, giving total losses of 6.56%. The most heavily
loaded elements of the system are the wind turbines. The generators representing the HVDC Light
connection are running at 48% and 47% at the offshore and onshore stations respectively (shown in
appendix 2).

Figure 49 shows the worst case load flow situation for the single sided ring design. The line between the
first wind turbine and the offshore substation is lost, meaning that the direction of the power flow is
reversed through the feeder. The most heavily loaded devices of the feeders are the wind turbines and the
wind turbine transformers, at 96%. All voltage levels are within their boundaries.


Figure 49 - Single sided ring, worst case operating conditions
The active power delivered to the grid is 504.2 MW, corresponding to 6.63% losses when comparing to
the nominal power output. The most heavily loaded elements of the system are the wind turbines and the
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 49 -

wind turbine transformers. The generators representing the HVDC Light connection are running at 48%
and 46% at the offshore and onshore stations respectively (shown in appendix 2).
7.5 Double sided ring
In each feeder of the double sided design, nine wind turbines are connected in series, and two feeders are
connected in parallel to provide redundancy. The wind turbines are rated at 5MW, with a power factor
limit of 0.95. The wind turbines are operating at 0.69kV. Each wind turbine is equipped with a 0.69/48kV
transformer, rated at 5.26 MVA. The cable ratings are 92.3 MVA throughout the loop, operating at 48 kV
due to the high power rating of the cable. The maximum rated voltage of the cable is 52 kV, or 1.083 pu.
Thus, all voltage levels lower than 1.083 pu are considered acceptable in the offshore grid. A value of 0.9
pu is used as the minimum acceptable voltage level.
7.5.1 AC/AC
Figure 50 shows the load flow situation of the double sided ring during normal operating conditions. The
most heavily loaded cables are running at 48%, since they are rated to provide redundancy. The most
heavily loaded elements of the feeders are the wind turbine transformers, which are rated at 96%. All
voltage levels are within their boundaries.


Figure 50 - Double sided ring, normal operating conditions
The active power delivered to the grid is 526.9 MW, giving total losses of 2.43%. The most heavily
loaded elements of the system are the transmission cables, rated at 100%.

Figure 51 shows the worst case load flow situation of the double sided ring-system. In this situation, one
of the lines connecting the first turbine of a feeder to the offshore substation is lost. The direction of the
power flow in this feeder is reversed, and all the turbines can continue to operate at full production. The
most heavily loaded cable now operates at 95%, and is the most heavily loaded device of the feeders,
together with the wind turbines which are also loaded at 95%. All voltage levels are within their
boundaries.


Figure 51 - Double sided ring, worst case operating conditions
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 50 -

The active power delivered to the grid is 526.3 MW, giving total losses of 2.54%. The most heavily
loaded elements of the system are the transmission cables, rated at 100%. The SVC must consume 0.1
MVAr in order to provide unity power factor in the PCC.
7.5.2 AC/DC
Figure 50 shows the load flow situation of the double sided ring during normal operating conditions.
None of the cables are rated above 48%, since they are rated to provide redundancy. The most heavily
loaded elements of the feeders are the wind turbine transformers, which are rated at 97%. All voltage
levels are within their boundaries.


Figure 52 - Double sided ring, normal operating conditions
The active power delivered to the grid is 505.0 MW, corresponding to 6.48% losses when comparing to
the nominal power output. The most heavily loaded elements of the system are the wind turbines and the
wind turbine transformers. The generators representing the HVDC Light connection are running at 48%
and 46% at the offshore and onshore stations respectively (shown in appendix 2).

Figure 53 shows the worst case load flow situation of the double sided ring-system. In this situation, one
of the lines connecting the first turbine of a feeder to the offshore substation is lost. The direction of the
power flow in this feeder is reversed, and no turbines need to be out of production. The most heavily
loaded cable is now running at 95% capacity. The most heavily loaded devices of the feeders are still the
wind turbines, at 97%. All voltage levels are within their boundaries.


Figure 53 - Double sided ring, worst case operating conditions
The active power delivered to the grid is 504.4 MW, corresponding to 6.59% losses when comparing to
the nominal power output. The most heavily loaded elements of the system are the wind turbines. The
generators representing the HVDC Light connection are running at 48% and 46% at the offshore and
onshore stations respectively (shown in appendix 2).
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 51 -

7.6 Shared ring
In each feeder in the shared ring design, nine wind turbines are connected in series, and four feeders share
a safety cable that provides redundancy. The wind turbines are rated at 5MW, with a power factor limit of
0.95. The wind turbines are operating at 0.69kV. Each wind turbine is equipped with a 0.69/33kV
transformer, rated at 5.26 MVA. The cable ratings are 48.6 MVA throughout the loop, operating at 33 kV.
The maximum rated voltage of the cable is 36 kV, or 1.091 pu. Thus, all voltage levels lower than 1.091
pu are considered acceptable in the offshore grid. The cables from the last wind turbine to the extra buses
at the feeder ends (see Figure 54) are modeled as lossless lines, representing the breaker system that is
used to rearrange the power flow if a cable is disconnected. A value of 0.9 pu is used as the minimum
acceptable voltage level.
7.6.1 AC/AC
Figure 54 shows the load flow situation for the shared ring design during normal operating conditions.
Due to the redundancy provided by the extra cable, no cables are rated higher than 76%. In the feeders, the
highest rated elements are the wind turbines, rated at 95%. All voltage levels are within their boundaries.


Figure 54 - Shared ring, normal operating conditions
The active power delivered to the grid is 525.5 MW, giving total losses of 2.69%. The most heavily
loaded elements of the system are the transmission cables, rated at 100%.

Figure 55 shows the worst case load flow situation for the shared ring design. The line between the first
wind turbine and the offshore substation of the feeder the furthest away from the redundant cable is lost,
meaning that the direction of the power flow is reversed through that feeder. To avoid overloading of the
cables at the other feeders, the connections from these to the redundant cable are disconnected, as
indicated by the dashed lines in Figure 55. The most heavily loaded cables are the ones connecting the
first wind turbine of each feeder to the offshore substation, and the redundant cable, all loaded at 92%.
The most heavily loaded devices of the power system are the wind turbines, which are running at 95%. All
voltage levels are within their boundaries.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 52 -


Figure 55 - Shared ring, worst case operating conditions
The active power delivered to the grid is 524.9 MW, giving total losses of 2.80%. The most heavily
loaded elements of the system are the transmission cables, rated at 100%. The SVC must consume 0.1
MVAr in order to provide unity power factor in the PCC.
7.6.2 AC/DC
Figure 56 shows the load flow situation for the shared ring design during normal operating conditions.
Due to the redundancy provided by the extra cable, no cables are rated higher than 77%. In the feeders, the
highest rated elements are the wind turbines, rated at 97%. All voltage levels are within their boundaries.


Figure 56 - Shared ring, normal operating conditions
The active power delivered to the grid is 502.4 MW, corresponding to 6.78% losses when comparing to
the nominal power output. The most heavily loaded elements of the system are the wind turbines and the
wind turbine transformers. The generators representing the HVDC Light connection are running at 48%
and 46% at the offshore and onshore stations respectively (shown in appendix 2).

Figure 57 shows the worst case load flow situation for the shared ring design. The line between the first
wind turbine and the offshore substation of the feeder the furthest away from the redundant cable is lost,
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 53 -

meaning that the direction of the power flow is reversed through that feeder. To avoid overloading of the
cables at the other feeders, the connections from these to the redundant cable are disconnected, as
indicated by the dashed lines in Figure 57. The most heavily loaded devices of the power system are the
wind turbines and the wind turbine transformers, running at 97% capacity. All voltage levels are within
their boundaries.

Figure 57 - Shared ring, worst case operating conditions
The active power delivered to the grid is 502.8 MW, corresponding to 6.89% losses when comparing to
the nominal power output. The most heavily loaded elements of the system are the wind turbines and the
wind turbine transformers. The generators representing the HVDC Light connection are running at 48%
and 46% at the offshore and onshore stations respectively (shown in appendix 2).
7.7 N-sided ring (n=4)
In each feeder the n-sided ring design, where n=4 in the simulations in this thesis, nine wind turbines are
connected in series, and four feeders are connected in parallel to provide redundancy. The wind turbines
are rated at 5MW, with a power factor limit of 0.95. The wind turbines are operating at 0.69kV. Each
wind turbine is equipped with a 0.69/33kV transformer, rated at 5.26 MVA. The cable ratings are 65.5
MVA throughout the loop, operating at 33 kV. The maximum rated voltage of the cable is 36 kV, or 1.091
pu. Thus, all voltage levels lower than 1.091 pu are considered acceptable in the offshore grid. A value of
0.9 pu is used as the minimum acceptable voltage level.
7.7.1 AC/AC
Figure 58 shows the n-sided ring design during normal operating conditions. Due to the redundancy
provided by the high cable rating (see equation 3-6), no cables are rated higher than 67%.The most heavily
loaded devices are the wind turbines and the wind turbine transformers, which are rated at 96%. All
voltage levels are within their boundaries.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 54 -


Figure 58 - N-sided ring, normal operating conditions
The active power delivered to the grid is 527.4 MW, giving total losses of 2.33%. The most heavily
loaded elements of the system are the transmission cables, rated at 100%.

Figure 59 shows the n-sided ring during worst case operating conditions. In this situation, a fault occurs at
one of the outermost feeders. This means that the direction of the power flow is reversed through the
feeder. The most heavily loaded cable is the one closest to the outermost feeder. The most heavily loaded
devices of the feeders are still the wind turbines and the wind turbine transformers, at 95%. All voltage
levels are within their boundaries.


Figure 59 - N-sided ring, worst case operating conditions
The active power delivered to the grid is 527.1 MW, giving total losses of 2.39%. The most heavily
loaded elements of the system are the transmission cables, rated at 100%. The SVC must consume 0.1
MVAr in order to provide unity power factor in the PCC.
7.7.2 AC/DC
Figure 60 shows the n-sided ring design during normal operating conditions. Due to the redundancy
provided by the high cable rating (see equation 3-6), no cables are rated higher than 68%.The most heavily
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 55 -

loaded devices are the wind turbines and the wind turbine transformers, which are rated at 96%. All
voltage levels are within their boundaries.


Figure 60 - N-sided ring, normal operating conditions
The active power delivered to the grid is 505.4 MW, corresponding to 6.41% losses when comparing to
the nominal power output. The most heavily loaded elements of the system are the wind turbines and the
wind turbine transformers. The generators representing the HVDC Light connection are running at 48%
and 46% at the offshore and onshore stations respectively (shown in appendix 2).

Figure 61 shows the n-sided ring during worst case operating conditions. In this situation, a fault occurs at
one of the outer feeders of the n feeders that are connected in parallel. This means that the direction of the
power flow is reversed through the feeder. The most heavily loaded cable is the one connecting the first
wind turbine of the feeder closest to the fault to the offshore substation. This is running at 93% of its full
capacity.The most heavily loaded devices of the feeders are still the wind turbines and the wind turbine
transformers, at 96%. All voltage levels are within their boundaries.


Figure 61 - N-sided ring, worst case operating conditions
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 56 -

The active power delivered to the grid is 504.7 MW, corresponding to 6.54% losses when comparing to
the nominal power output. The most heavily loaded elements of the system are the wind turbines and the
wind turbine transformers. The generators representing the HVDC Light connection are running at 48%
and 46% at the offshore and onshore stations respectively (shown in appendix 2).
7.8 Star design
In the star design, nine wind turbines are connected in a star to the same connection point. From this point,
the power is transmitted in a common cable to the offshore substation. The wind turbines are rated at
5MW, with a power factor limit of 0.95. The wind turbines are operating at 0.69kV. Each wind turbine is
equipped with a 0.69/11kV transformer, rated at 5.26 MVA. The cable ratings are 5.5 MVA, operating at
11 kV. The maximum rated voltage of the cable is 12 kV, or 1.091 pu. Thus, all voltage levels lower than
1.091 pu are considered acceptable in the star. From the centre bus, the voltage is transformed from 11kV
to 33kV. The cable rating of the cable between the centre bus of the star and the offshore substation is
48.6 MVA, rated at 33 kV. The maximum rated voltage of the cable is 36 kV, or 1.091 pu. Thus, for the
transmission from the star to the offshore substation, a voltage level lower than 1.091 pu are considered
acceptable in the offshore grid. A value of 0.9 pu is used as the minimum acceptable voltage level.
7.8.1 AC/AC
Figure 62 shows the star design during normal operating conditions. The 11 kV cables are loaded at 90%
and the 33 kV cable is loaded at 95%. The most heavily loaded devices in the system are the wind turbines
and the33 kV cable connecting the star and the substation, being loaded at 95%. All voltage levels are
within their boundaries.


Figure 62 - Star design, normal operating conditions
The active power delivered to the grid is 517.5 MW, giving total losses of 4.17%. The most heavily
loaded elements of the system are the transmission cables, rated at 100%.

Figure 63 shows the load flow result during worst case operating conditions, that is, when one star is out
of service. The only difference in the running conditions is that the star draws slightly more reactive
power. Except from that, the devices are running at the same load levels as earlier. All voltage levels are
within their boundaries.

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 57 -


Figure 63 - Star design, worst case operating conditions
The active power delivered to the grid is 474.4 MW, giving total losses of 12.15% when comparing to the
nominal power output. The most heavily loaded elements of the system are the transmission cables, rated
at 100%. The SVC must consume 6.1 MVAr in order to provide unity power factor in the PCC.
7.8.2 AC/DC
Figure 64 shows the star design during normal operating conditions. The 11 kV cables are loaded at 93%
and the 33kV connection to the offshore substation is running at 95%. The most heavily loaded devices in
the system are the wind turbines and the wind turbine transformers, running at 99%. All voltage levels are
within their boundaries.


Figure 64 - Star design, normal operating conditions
The active power delivered to the grid is 495.8 MW, corresponding to 8.79% losses when comparing to
the nominal power output. The most heavily loaded elements of the system are the wind turbines and the
wind turbine transformers. The generators representing the HVDC Light connection are running at 48%
and 46% at the offshore and onshore stations respectively (shown in appendix 2).

Figure 65 shows the load flow result during worst case operating conditions, that is, when one star is out
of service. The 11kV cables now run at 92%, and the wind turbine transformers are running at 98%. The
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 7
Load flow simulations

- 58 -

most heavily loaded elements of the star are still the wind turbines, running at 99%. All voltage levels are
within their boundaries.


Figure 65 - Star design, DC transmission, worst case operating conditions
The active power delivered to the grid is 453.8 MW, corresponding to 15.96% losses when comparing to
the nominal power output. The most heavily loaded elements of the system are the wind turbines. The
generators representing the HVDC Light connection are running at 44% and 42% at the offshore and
onshore stations respectively (shown in appendix 2).

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 59 -

8 Dynamic simulations
This chapter describes the dynamic simulations that have been executed as a part of the work with this
thesis. In chapter 3.5, six options for the design of the offshore grid were presented. In this chapter, their
dynamic behavior is studied and compared. In addition, the impact of the choice of transmission
technology from the wind farm to the PCC is studied. Three fault scenarios are presented. This means that
a total of 36 dynamic simulations are executed. For all wind farm layouts, the active power P, the reactive
power Q, the generator speed and the voltage U are plotted at selected generator buses. It has been
chosen to study the rotor speed instead of the rotor angle, since the DFIG can control the rotor angle in the
partial scale frequency controller. (see Figure 5) Only the most relevant plots are reproduced in this
chapter, but all plots can be found in appendix 5.

All dynamic simulations are based on the normal operating conditions-load flow situation, as described
in chapter 7. For all three fault scenarios, the fault applied to the system is a three phase short circuit to
ground, lasting 150 ms before the circuit breakers on the faulted cable open, and the cable is
disconnected.

When comparing the wind farm layouts, two main factors are studied; how the grid design influences the
dynamic behavior, and how the transmission choice influences the dynamic behavior. This is done by
studying the rotor speed stability and voltage stability in the system. The maximum oscillation and time it
takes for the rotor speed to stabilize is studied, as is the voltage response in terms of maximum voltage dip
and the maximum overvoltage preceding a fault. As the initial rotor speed of all the wind turbines of all
the systems are similar, at 0.2 pu, the oscillations are studied by looking at the per unit values of the
deviation from the initial value.

The voltage level is case specific and also varies from turbine to turbine. In order to compare the voltage
dips, the value has been calculated as percentage of the original voltage level at the bus in question. The
original voltage level values are gathered from the normal operating conditions-load flow cases.
8.1 Fault scenario one
In fault scenario one, a fault is applied to one of the two transmission lines between the offshore wind
farm and the PCC, as illustrated in Figure 66:


Figure 66 - Fault scenario one
This scenario is studied to investigate two issues:

1) How does the offshore wind farm layout influence the dynamic behavior following a transmission
fault?
2) How does the transmission choice influence the dynamic behavior of the wind turbines following
a transmission fault?

Offshore wind
farm
Onshore power
system
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 60 -

A fault in the transmission system will have the largest impact on the generators closest to the fault, i.e.
the wind turbine generators situated at the buses that lie the closest to the offshore substation. Therefore,
only the dynamic behavior of the generator at one of these buses is studied. Figure 67 shows this plot for
the radial design. For the five other designs, the dynamic behavior is similar. Thus, their dynamic response
is not shown here, but can be seen in appendix 5.


Figure 67 Fault scenario one. Radial design. P (green), Q (blue), (light blue) and U (purple) of the wind turbine
generator closest to the substation. Top: AC/AC Bottom: AC/DC.
The dynamic response is clearly stable, with the oscillations being fully damped at approximately seven
seconds for the AC/AC system and eight seconds for the AC/DC system. The minimum and maximum
deviations of the twelve designs are compared below:

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 61 -

For fault scenario one, Figure 68 shows the maximum and minimum deviation of the generator speed
and voltage level U at the bus closest to the offshore substation for all 12 designs.


Figure 68 - Fault scenario one. Maximum speed deviation [pu on system base] and voltage deviation [% of initial value] )
of the wind turbine generator closest to the substation.
:
The choice of offshore grid does not have a large impact on the maximum and minimum speed deviation.
Only small differences are observable.
The impact of the choice of transmission system is clearly of importance. For all designs, the speed
deviation is clearly and consequently higher for the AC/DC systems than for the AC/AC systems.

-0,0568 -0,0569 -0,0571 -0,0569 -0,0569
-0,0550
-0,0646 -0,0637 -0,0645 -0,0642 -0,0644
-0,0658
0,0609 0,0611 0,0611 0,0609 0,0616 0,0606
0,0690 0,0682 0,0691 0,0686 0,0688
0,0724
-0,0800
-0,0600
-0,0400
-0,0200
0,0000
0,0200
0,0400
0,0600
0,0800
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
-72,8 %
-74,8 % -74,8 %
-73,7 % -74,0 %
-63,3 %
-77,5 %
-78,4 % -78,7 %
-77,8 % -77,7 %
-69,8 %
7,4 % 7,6 % 7,5 % 7,4 % 7,5 % 7,1 %
2,4 % 2,7 %
5,1 %
2,8 % 3,1 %
1,6 %
-90,0 %
-80,0 %
-70,0 %
-60,0 %
-50,0 %
-40,0 %
-30,0 %
-20,0 %
-10,0 %
0,0 %
10,0 %
20,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 62 -

U:
The star design separates clearly from the other designs. The voltage drop is lower than for the rest of the
designs, and the maximum voltage is also smaller. The other designs show small variations in the voltage
deviation.
The impact of the choice of transmission system is clearly of importance. For all designs, the voltage drop
is larger for DC transmission than for AC transmission. Nevertheless, the maximum values of the voltages
are clearly and consequently smaller for the AC/DC systems than the AC/AC systems.
8.2 Fault scenario two
In fault scenario two, a fault is applied to the cable connecting the innermost turbine of one of the feeders
to the offshore substation. For the radial and star design, this means that the entire feeder is lost as a
consequence of this fault, since no redundancy is provided. This is illustrated in Figure 69:


Figure 69 - Fault scenario two
For fault scenario two, the dynamic response at two buses is plotted. The scope is to investigate how a
fault of type two affects:

1) The buses in the same feeder as the one subject to the fault
2) The buses in the rest of the offshore wind farm.

The first bus to be studied is the bus the closest to the fault, i.e. the innermost bus of the feeder, when seen
from the offshore substation. For the radial and star design, when the feeder is disconnected it is isolated
from the rest of the network, and the dynamic behavior in the isolated feeder is unstable. For the
redundant designs, this problem is avoided. This is illustrated in Figure 70 and Figure 71:
Connection to collector hub
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 63 -



Figure 70 Fault scenario two. Radial design. P (green), Q (blue), (light blue) and U (purple) of the wind turbine
generator closest to the substation in the same feeder as the fault occurs. Top: AC/AC Bottom: AC/DC.
The rotor speed does not return to its original value and there is a drop in the active and reactive power.
The instability occurs faster for the AC/DC system than for the AC/AC system. The voltage increases
throughout the first ten seconds after the fault. For the redundant designs, these post-fault problems are
avoided. Figure 71 shows the dynamic behavior of the wind turbine generator closest to the substation in
the feeder that is isolated when the shared ring concept is applied to the offshore grid:
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 64 -



Figure 71 Fault scenario two. Shared ring design. P (green), Q (blue), (light blue) and U (purple) of the wind turbine
generator closest to the substation when the fault occurs in the same feeder. Top: AC/AC Bottom: AC/DC.
The figure shows that all values return to their original values after approximately seven seconds. The
dynamic behavior of the different wind farm layout options are compared in Figure 72:

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 65 -



Figure 72 Fault scenario two. Maximum speed deviation [pu on system base] and voltage deviation [% of initial value] )
of the wind turbine generator closest to the substation when the fault occurs in the same feeder as the generator is placed.
:
The choice of offshore electrical collector does not affect the maximum and minimum speed deviation
considerably. The speed deviation at the radial and star design is not considered, as the speed is clearly
unstable at these buses.
The impact of the choice of transmission system is of importance even though the differences are quite
small, the AC/DC system consequently gives higher speed deviations than the AC/AC system.

U:
There is no big difference in the voltage drop at the layouts that does not lead to unstable post-fault
operation. The maximum voltage is clearly highest at the two buses where unstable operation is obtained
after the fault. The star design gives the lowest voltage drop, but clearly the highest voltage rise for the
AC/AC system. For the AC/DC system, the radial design gives the highest voltage rise, while the star
-0,0583 -0,0585 -0,0583 -0,0587 -0,0589
-0,0606
-0,0595 -0,0599
0,0624 0,0625 0,0623 0,0628
0,0646
0,0660
0,0650 0,0653
-0,0800
-0,0600
-0,0400
-0,0200
0,0000
0,0200
0,0400
0,0600
0,0800
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
-86,5 % -86,7 % -86,1 % -86,5 % -86,6 %
-74,8 %
-89,7 % -89,5 % -89,5 % -89,5 % -89,5 %
-80,8 %
27,9 %
4,7 %
6,9 %
4,7 % 4,2 %
32,7 %
14,6 %
-0,1 %
5,1 %
1,4 % 1,2 %
11,6 %
-100,0 %
-80,0 %
-60,0 %
-40,0 %
-20,0 %
0,0 %
20,0 %
40,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 66 -

design still gives the smallest drop. For all cases, the AC/DC system gives larger drops and smaller
maximum values than for the AC/AC systems.

The second bus to be studied is the first bus of one of the feeders in the wind farm that was not subject to
the fault. The response of the bus that is directly connected to the offshore substation is shown in Figure
73:


Figure 73 Fault scenario two. Radial design. P (green), Q (blue), (light blue) and U (purple) of the wind turbine
generator closest to the substation when the fault occurs in one of the other feeders. Top: AC/AC Bottom: AC/DC.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 67 -

All dynamic responses are clearly stable. It takes approximately five seconds for the oscillations to be
fully damped. The voltage drop and reactive power rise are relatively small.

The minimum and maximum deviations of the twelve designs at the relevant buses are compared below:

For fault scenario two, Figure 74 shows the maximum and minimum deviation of the generator speed
and voltage level U at the bus closest to the offshore substation in one of the other feeders for all 12
designs:

Figure 74 Maximum speed deviation [pu on system base] and voltage deviation [% of initial value] ) of the wind turbine
generator closest to the substation following fault 2 when the fault occurs in a different feeder than where the generator is
placed.
:
The choice of offshore electrical collector clearly influences the maximum and minimum speed deviation.
The radial design gives the smallest deviations, with the star design as number two, giving slightly higher
oscillations. The rest of the order is as follows: Single sided, shared, n-sided, double sided.
-0,0114
-0,0169
-0,0270
-0,0209
-0,0242
-0,0119
-0,0100
-0,0158
-0,0296
-0,0216
-0,0248
-0,0113
0,0104
0,0169
0,0275
0,0211
0,0247
0,0110
0,0097
0,0163
0,0307
0,0222
0,0256
0,0109
-0,0400
-0,0300
-0,0200
-0,0100
0,0000
0,0100
0,0200
0,0300
0,0400
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
-27,2 %
-35,0 %
-48,6 %
-40,7 %
-43,7 %
-26,2 % -26,1 %
-33,2 %
-50,6 %
-40,5 %
-43,6 %
-26,8 %
5,5 %
5,0 %
6,8 %
4,8 % 4,7 % 4,9 %
0,2 %
-1,1 %
3,6 %
0,5 % 0,6 % 0,5 %
-60,0 %
-50,0 %
-40,0 %
-30,0 %
-20,0 %
-10,0 %
0,0 %
10,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 68 -

The impact of the choice of transmission system is also clearly of importance. For the star and radial
design, the oscillations are smaller for the AC/DC system than for the AC/AC system. For the rest of the
systems, the AC/DC system gives larger oscillations than for the AC/AC system

U:
The star and radial designs clearly stand out as the ones giving the smallest voltage drops. Using AC
transmission, the star design gives the smallest voltage deviations, while the radial design gives the
smallest deviations when using DC transmission. The double sided design clearly gives the largest drop in
the voltage.
The impact of the choice of transmission system varies when it comes to the voltage drops, but all AC/DC
layouts consequently give smaller maximum values of the voltage.
8.3 Fault scenario three
In fault scenario three, a fault is applied in the outermost cable of the feeder. For the radial and looped
designs, this means that the fault occurs between the eighth and ninth turbine. For the star design, the fault
occurs between the centre bus and one of buses at the end of one of the arms of the star. This is
illustrated in Figure 75:


Figure 75 - Fault scenario three
For fault scenario three, the dynamic response at three buses is plotted. The scope is to investigate how a
fault of type two affects the following buses:

1) The bus next to the fault that is further away from the offshore substation than the fault (the bus
right of the fault in the left drawing, and the lower right bus of the right drawing, of Figure 75).
2) The bus next to the fault that is closer to the substation than the fault (the bus left of the fault in
the left drawing, and the central bus of the right drawing, of Figure 75)
3) The bus the closest to the offshore substation in the same feeder as the fault occurs.

As the right drawing of Figure 75 illustrates, for the star design, the buses described in point 2) and 3) is
the same, i.e. the central bus of the star.

The dynamic response for the bus described in point 1) is shown for the radial design in Figure 76:
Connection to collector hub
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 69 -



Figure 76 - Fault scenario three. Radial design. P (green), Q (blue), (light blue) and U (purple) of the wind turbine
generator further away from the offshore substation than the fault point. Top: AC/AC Bottom: AC/DC.
The figures show that the response is clearly unstable. The instability occurs faster for the AC/DC system
than for the AC/AC system. This instability is the case for the radial and star design. For the redundant
designs, this problem is avoided. The dynamic response of the same bus is shown for the shared design
below:

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 70 -


Figure 77 - Fault scenario three. Shared ring design. P (green), Q (blue), (light blue) and U (purple) of the wind turbine
generator further away from the offshore substation than the fault point. Top: AC/AC Bottom: AC/DC.
The looped design of the shared ring provides stability after the fault. The oscillations are fully damped
after approximately seven seconds. The impact of the wind farm layout on the dynamic response is shown
in Figure 78:
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 71 -


Figure 78 - Fault scenario three. Maximum speed deviation [pu on system base] and voltage deviation [% of initial value] )
of the wind turbine generator furthest away from the substation when the fault occurs in the same feeder as the generator
is placed.
:
The choice of offshore electrical collector affects the maximum and minimum speed deviation
considerably. The smallest deviation is for the shared design, and the largest for the doubly sided design,
both for the AC/AC systems and the AC/DC systems. The speed deviation at the radial and star design is
not considered, as the speed is clearly unstable at these buses.
The impact of the choice of transmission system is of importance. The AC/DC systems consequently give
higher speed deviations than the AC/AC system.

-0,0552
-0,0562
-0,0523
-0,0548
-0,0560
-0,0586
-0,0544
-0,0563
0,0589
0,0599
0,0552
0,0582
0,0609
0,0636
0,0586
0,0611
-0,0800
-0,0600
-0,0400
-0,0200
0,0000
0,0200
0,0400
0,0600
0,0800
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
-86,4 %
-84,5 % -84,3 %
-81,3 %
-83,3 %
-85,4 %
-89,7 %
-88,5 % -89,0 %
-85,7 %
-87,5 %
-89,5 %
27,8 %
5,1 %
6,9 %
4,6 % 4,7 %
31,3 %
14,1 %
-0,5 %
4,0 %
0,8 % 0,9 %
10,4 %
-100,0 %
-80,0 %
-60,0 %
-40,0 %
-20,0 %
0,0 %
20,0 %
40,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 72 -

U:
There is a noticeable difference in the voltage drop at the layouts that does not lead to unstable post-fault
operation. The smallest deviation is for the shared design, and the largest for the doubly sided design, both
for the AC/AC systems and the AC/DC systems. The maximum voltage is clearly highest at the two buses
where unstable operation is obtained after the fault. The star design gives the lowest voltage drop, but
clearly the highest voltage rise for the AC/AC system. For the AC/DC system, the radial design gives the
highest voltage rise, while the star design still gives the smallest drop. For all cases, the AC/DC system
gives larger drops and smaller maximum values than for the AC/AC systems.

The dynamic response for the bus next to the fault, that is closer to the substation than the fault, is shown
for the radial design in Figure 79:


Figure 79 Fault scenario three. Radial design. P (green), Q (blue), (light blue) and U (purple) of the wind turbine
generator closer to the offshore substation than the fault point. Top: AC/AC Bottom: AC/DC.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 73 -

The oscillations in Figure 79 are fully damped after approximately seven seconds, and the post fault
response is clearly stable. For the AC/AC system, the reactive power is negatively increasing until it
stabilizes at eight seconds. For the AC/DC design, it oscillates and stabilizes at the pre-fault value.

The dynamic response for the bus closest to the offshore substation, in the same feeder as the fault occurs,
is shown for the radial design in Figure 80:



Figure 80 - Fault scenario three. Radial design. P (green), Q (blue), (light blue) and U (purple) of the wind turbine
generator closest to the offshore substation. Top: AC/AC Bottom: AC/DC.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 74 -

As expected, the oscillations are smaller, and the time it takes for the oscillations to be fully damped is
smaller, approximately six seconds. The reactive power also reaches a stable operating point faster than
for the wind turbine further out in the radial.

Finally, the dynamic response of the centre bus of the star design is shown in Figure 81:



Figure 81 - Fault scenario three. Star design. P (green), Q (blue), (light blue) and U (purple) of the wind turbine
generator at the central bus of the star. Top: AC/AC Bottom: AC/DC.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 75 -

The oscillations are fully damped after approximately seven seconds, and the post fault response is clearly
stable. For the AC/AC system, the reactive power is negatively increasing all the way up to ten seconds.
For the AC/DC design, it oscillates and stabilizes at the pre-fault value.

The minimum and maximum deviations of the twelve designs at the relevant buses are compared below:
For fault scenario three, Figure 82 shows the maximum and minimum deviation of the generator speed
and voltage level U at the bus next to the fault that is closer to the substation than the fault (the bus left
of the fault in the left drawing, and the central bus of the right drawing, of Figure 75)

Figure 82 - Fault scenario three. Maximum speed deviation [pu on system base] and voltage deviation [% of initial value] )
of the wind turbine generator bus next to the fault, that is closer to the substation than the fault.
:
The choice of offshore grid has a small impact on the speed deviations.
The impact of the choice of transmission system is of larger importance. For all designs, the speed
deviation is higher for the AC/DC systems than for the AC/AC systems.

-0,0573 -0,0579 -0,0580 -0,0586 -0,0586 -0,0577
-0,0590 -0,0585
-0,0601 -0,0599 -0,0597 -0,0602
0,0615 0,0621 0,0621 0,0626 0,0626 0,0618
0,0638 0,0640
0,0654 0,0654 0,0651 0,0651
-0,0800
-0,0600
-0,0400
-0,0200
0,0000
0,0200
0,0400
0,0600
0,0800
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
-86,5 % -86,7 % -86,1 % -86,5 % -86,6 % -86,0 %
-89,7 % -89,5 % -89,5 % -89,6 % -89,5 % -89,9 %
3,4 %
4,9 %
6,9 %
4,6 % 4,7 %
1,3 % 0,7 %
-0,4 %
3,9 %
0,8 % 0,9 % 0,2 %
-100,0 %
-80,0 %
-60,0 %
-40,0 %
-20,0 %
0,0 %
20,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 76 -

U:
The choice of offshore grid has a small impact on the voltage drops, but impacts the voltage rise. The
double sided design has the highest rise in the voltage following the fault.
The impact of the choice of transmission system is of larger importance. For all designs, the voltage drop
is larger for the AC/DC systems than for the AC/AC systems, while the maximum values of the voltage
are clearly smaller.

For fault scenario three, Figure 83 shows the maximum and minimum deviation of the generator speed
and voltage level U at the bus closest to the offshore substation in the same feeder as the one where the
fault occurs for all 12 designs:

Figure 83 - Fault scenario three. Maximum speed deviation [pu on system base] and voltage deviation [% of initial value] )
of the wind turbine generator bus closest to the offshore substation in the same feeder as the fault occurs.


-0,0352
-0,0395
-0,0431
-0,0450 -0,0442
-0,0577
-0,0374
-0,0409
-0,0464 -0,0472
-0,0462
-0,0602
0,0365
0,0411
0,0448
0,0468 0,0460
0,0618
0,0388
0,0432
0,0491
0,0504
0,0490
0,0651
-0,0800
-0,0600
-0,0400
-0,0200
0,0000
0,0200
0,0400
0,0600
0,0800
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
-59,7 %
-65,5 %
-70,4 %
-73,2 %
-71,3 %
-86,0 %
-63,0 %
-68,1 %
-74,7 %
-76,7 %
-74,6 %
-89,9 %
3,6 %
5,0 %
6,9 %
4,7 % 4,8 %
1,3 %
0,5 %
-0,6 %
3,7 %
0,7 % 0,7 % 0,2 %
-100,0 %
-80,0 %
-60,0 %
-40,0 %
-20,0 %
0,0 %
20,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 8
Dynamic simulations

- 77 -

:
The choice of offshore electrical collector clearly influences the maximum and minimum speed deviation.
The radial design gives the smallest deviations, while the star gives the largest deviations.
The impact of the choice of transmission system is also of importance. The AC/DC systems consequently
give larger deviations than the AC/DC systems.

U:
The choice of offshore electrical collector clearly influences the maximum and minimum voltage
deviation as well. The radial design gives the smallest deviations, while the star gives the largest
deviations, just as for the speed.
The impact of the choice of transmission system is also of importance. The AC/DC systems consequently
give larger voltage drops than the AC/DC systems, while the maximum values of the voltage are clearly
smaller.


Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 9
Discussion

- 78 -

9 Discussion
In this chapter, the most important results from the load flow simulations and dynamic simulations are
presented and discussed. For the load flow situation, this is done mainly by comparing the system losses at
the two operating situations. For the dynamic simulations, this is done by comparing the behavior of the
9.1 Load flow simulations
Figure 84 shows that the losses are considerably lower in the AC/AC layouts than in the AC/DC layouts.
For the AC/AC systems, the losses vary from 2,33% to 4,17% of the total power production, while for the
AC/DC layouts, this percentage is higher, varying between 6,41% and 8,19%.



Figure 84 - Active power losses for the different wind farm layouts. The top two diagrams show MW losses, the bottom
two show losses as percentage of the total produced power in the wind farm.
During normal operating conditions, the difference between the AC/DC system losses and the AC/AC
system losses varies from 22.1 MW for the shared ring design to 21.7 MW for the star design.

The layout giving the smallest losses, both for AC and DC transmission to shore, is the n-sided ring
design, giving total losses of 12.6MW (2.33%) for the AC/AC system, and 34.6MW (6.41%) for the
AC/DC system. The layout giving the highest losses is the star system, giving 46-79% higher losses than
the other designs for the AC/AC layout, and 19-28% higher losses than the other designs for the AC/DC
layout.

By comparing the top right and top left diagram of Figure 84, it is clear that the radial and star designs are
the ones giving the highest increase in the active power losses when the operating conditions change from
15,4
37,3
13,4
35,4
13,1
35
14,5
36,6
12,6
34,6
22,5
44,2
0
5
10
15
20
25
30
35
40
45
50
AC DC AC DC AC DC AC DC AC DC AC DC
RAD SS DS Shared NS Star
During normal operation
59,1
79,8
13,8
35,8
13,7
35,6
15,1
37,2
12,9
35,3
65,6
86,2
0
10
20
30
40
50
60
70
80
90
100
AC DC AC DC AC DC AC DC AC DC AC DC
RAD SS DS Shared NS Star
During worst case operation
2,85 %
6,91 %
2,48 %
6,56 %
2,43 %
6,48 %
2,69 %
6,78 %
2,33 %
6,41 %
4,17 %
8,19 %
0,00 %
1,00 %
2,00 %
3,00 %
4,00 %
5,00 %
6,00 %
7,00 %
8,00 %
9,00 %
AC DC AC DC AC DC AC DC AC DC AC DC
RAD SS DS Shared NS Star
During normal operation
10,94 %
14,78 %
2,56 %
6,63 %
2,54 %
6,59 %
2,80 %
6,89 %
2,39 %
6,54 %
12,15 %
15,96 %
0,00 %
2,00 %
4,00 %
6,00 %
8,00 %
10,00 %
12,00 %
14,00 %
16,00 %
18,00 %
AC DC AC DC AC DC AC DC AC DC AC DC
RAD SS DS Shared NS Star
During worst case operation
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 9
Discussion

- 79 -

normal to worst case. This is because the worst case running conditions for these two systems imply the
loss of production from nine wind turbines, since no redundancy is provided. The difference for the radial
design is 43.7MW for the AC/AC system and 42.5MW for the AC/DC system. For the star design, the
difference is 43.1MW for the AC/AC system and 42.0MW for the AC/DC system.

For the rest of the wind farm layouts, the differences in power losses are small when comparing the
normal operating conditions to the worst case operating conditions, ranging from 0.3MW to 0.7MW.

It should be noted that the AC transmission cables are running at 100% of their power ratings. This is due
to the high capacitive charging currents produced in the cables. Also, in the AC/AC layouts, there was a
need for fixed reactors at a total size of approximately 1600 MVAr (!) for all cases, larger than any
produced shunt reactor in the world today. In addition, an SVC of minimum 175 MVArs had to be
installed in order to fulfill the grid codes regarding power factor control of the wind farm output power.
This confirms what was written in the introduction, where it was stated that long AC cable connections to
shore can lead to the need of very large and expensive reactive power compensators.

If the distance were to increase further, the cable rating would be violated, and yet another AC cable
would have to be added to the transmission to be able to transport the power to shore. This would lead to
the need for yet another shunt reactor in the several hundreds of megawatts range. The SVC would not
have to be increased, as it is based on the active power transmitted, and a maximum power factor of 0.95.
No economical calculations have been conducted in this thesis, but based on the amount of extra
equipment needed for an AC connection with three cables it is likely that the HVDC Light link would give
considerably better profitability. With no calculations to back this assertion, this is only speculations. The
issue should be further looked into by conducting a feasibility study prior to the construction of the wind
farm.

The load flow results describe the consequences of a fault, but says nothing about the probability of a fault
occurring. This comparison does not take into consideration that the probability for faults may decrease as
the cable ratings increase. Instead, the best case and worst case cable failure rate numbers from chapter 5.3
are used as a basis for the calculations. Based on those failure rates, the number of failures for each design
is calculated as,:

wf
cable
faults
LT
l
CFR N =
100

9-1
, where N
fault
denotes the number of faults per year and:

LT
wf
= Life time of the wind farm [years]
l
cable
= length of cable [km]
CFR = cable failure rate [failures/100km/year]

Based on a wind farm lifetime of 30 years, the expected numbers of faults for each design are shown in
Table 7 and Table 8:

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 9
Discussion

- 80 -

Cable \design: Radial Single sided Double sided Shared N-sided Star
WT1-Offshore substation 1,4 1,4 1,4 1,4 1,4 1,4
WT1-WT2 0,1 0,1 0,1 0,1 0,1 0,1
WT2-WT3 0,1 0,1 0,1 0,1 0,1 0,1
WT3-WT4 0,1 0,1 0,1 0,1 0,1 0,3
WT4-WT5 0,1 0,1 0,1 0,1 0,1 0,3
WT5-WT6 0,1 0,1 0,1 0,1 0,1 0,3
WT6-WT7 0,1 0,1 0,1 0,1 0,1 0,3
WT7-WT8 0,1 0,1 0,1 0,1 0,1 0,3
WT8-WT9 0,1 0,1 0,1 0,1 0,1 0,3
WT9-Offshore substation 0 2,6 0 2,6 0 0
Feeder to feeder 0 0 0,1 0,1 0,1 0
In total 2,6 5,2 2,7 5,3 2,7 3,4
Table 7 - Expected number of failures during the lifetime of the wind farm (LT = 30 years, CFR = 0.08)
Cable \design: Radial Single sided Double sided Shared N-sided Star
WT1-Offshore substation 3,6 3,6 3,6 3,6 3,6 3,6
WT1-WT2 0,4 0,4 0,4 0,4 0,4 0,4
WT2-WT3 0,4 0,4 0,4 0,4 0,4 0,4
WT3-WT4 0,4 0,4 0,4 0,4 0,4 0,6
WT4-WT5 0,4 0,4 0,4 0,4 0,4 0,6
WT5-WT6 0,4 0,4 0,4 0,4 0,4 0,7
WT6-WT7 0,4 0,4 0,4 0,4 0,4 0,7
WT7-WT8 0,4 0,4 0,4 0,4 0,4 0,7
WT8-WT9 0,4 0,4 0,4 0,4 0,4 0,7
WT9-Offshore substation 0 6,5 0 6,5 0 0
Feeder to feeder 0 0 0,2 0,4 0,3 0
In total 6,5 13 6,7 13,3 6,8 8,6
Table 8 - Expected number of failures during the lifetime of the wind farm (LT= 30 years, CFR = 0.20)
The lowest number of expected faults can be found at the radial design, while the shared design gives the
highest amount of expected faults. Nevertheless, the consequences of losing the lines vary. Table 9 shows
the extra power losses in the system as a consequence of a failure in one of the cables:

P
loss,extra
[MW]
Cable \design:
Radial Single sided Double sided Shared N-sided Star
WT1-Offshore substation 43,7 0,4 0,6 0,6 0,3 43,1
WT1-WT2 38,8 0,2 0,5 0,4 0,2 4,7
WT2-WT3 33,9 0,1 0,4 0,3 0,2 4,7
WT3-WT4 29 0 0,3 0,2 0,1 4,7
WT4-WT5 24,1 0 0,2 0,2 0,1 4,7
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 9
Discussion

- 81 -

WT5-WT6 19,3 -0,1 0,1 0,1 0 4,7
WT6-WT7 14,4 -0,1 0,1 0,1 0 4,7
WT7-WT8 9,6 0 0 0 0 4,7
WT8-WT9 4,8 0 0 0 0 4,7
WT9-Offshore substation 0,1 0,1
Feeder to feeder 0,3 0
Table 9 Extra power losses compared to normal operating conditions, when a cable is lost.
The table shows that the consequences of losing a line in terms of power losses is a lot bigger for the
radial design than for the others. The star design is a clear number two, while the looped designs are
clearly superior when it comes to providing redundancy. When knowing the mean time to repair of a cable
offshore, and having a good forecast for the electricity price, the expected loss of income for the wind
farm due to non-provided redundancy can be calculated. Since this will only be speculations, and are
beyond the scope of this thesis, this will not be done here.

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 9
Discussion

- 82 -

9.2 Dynamic simulations
For the dynamic simulations, Table 11 shows how the maximum speed deviations are dependent on
transmission technology and offshore grid design:

SPEED DEVIATIONS FAULT 1 FAULT 2 FAULT 3
First bus of
neighboring
feeder
Transmission
DC > AC

DC>AC except for
star and radial.

Not studied
Design
Small differences

DS>NS>Shared>SS>
Star>Rad
for both AC and DC
Not studied
First bus of
same feeder
Transmission
Not studied DC > AC. DC > AC

Design
Not studied Rad and star
unstable.
Small differences
AC:
Star>Shared>NS>DS>SS>Rad
DC:
Star>Shared>DS>NS>SS>Rad
Bus closest to
the fault and
closer to the
substation
Transmission
Not studied Not studied DC > AC
Design
Not studied Not studied Small differences
Bus closest to
the fault,
further away
from the
substation
Transmission
Not studied Not studied DC > AC
Design
Not studied Not studied Rad and star unstable
AC:DS>SS>NS>SH
DC: DS>NS>SS>SH

Table 10 - Comparison of the speed deviation for dynamic simulations. A>B means that the maximum deviation from the
steady state value is larger for A than for B, i.e. the response is worse.
For all cases except two, the AC/DC layouts give larger maximum speed deviation from the steady state
value during a fault. The two exceptions are found at the first bus of the generator at the neighboring
feeder to the fault, when using the star and radial design during fault scenario two.

The speed deviations are also consequently larger when using HVDC Light transmission compared to
using AC cable transmission, with two exceptions, as described in the beginning of this chapter. The
reason why the speed deviations are bigger for the DC systems than for the AC systems can be explained
by looking at equation 2-8. The offshore electric power is plotted in the dynamic simulations chapter.
Diagrams showing the maximum and minimum values of the electric power can be found in appendix 5.
The diagrams show that for all cases, the electric power during the fault is larger for AC transmission than
for HVDC Light transmission, P
E,AC
>P
E,DC
. during the fault. From equation 2-8, it can be seen that when
the electrical power P
E
increases, the rotor speed change decreases. This explains to some degree why the
maximum speed deviation is larger for the DC systems than for the AC systems. The above reasoning
only considers the rotor speed change during the three phase fault, and thus only describes the rotor speed
change during the fault. It does not take into consideration the damping capability of the system. When
conducting the DC simulations, the control parameter VLTFLG is equal to 1 for the AC/DC simulations,
and 0 for the AC/AC simulations, in order to get the initial conditions check to be OK. This change in the
wind turbine controller settings will affect the dynamic behavior, and might be the explanation to why the
rotor speed deviation is smaller for the AC/AC system than the AC/DC system.

Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 9
Discussion

- 83 -

When looking at how the design affects the speed response, it is difficult to range the designs from best to
worse, since different faults give different results. What can be noted is that for fault scenario three, the
star design gives the highest deviation for the first bus of the same feeder, since this bus is directly
coupled to the fault. The different systems have different cables, with different cable parameters. The
dynamic response depends on the total impedance between the generator and fault, and with different
impedances comes different dynamic responses. The star and radial designs are the only ones giving
unstable operation, since the fault isolates certain generators. All other generators in all systems have a
very stable operation.

Table 11 shows how the voltage drop and rise is dependent on transmission technology and offshore grid
design:

VOLTAGE DEVIATIONS FAULT 1 FAULT 2 FAULT 3
First bus of
neighboring
feeder
Transmission
Drop: DC > AC
Rise: AC > DC

Drop: DC > AC
Rise: AC > DC

Not studied
Design
Star clearly lowest.
AC:
DS=SS>NS>Shared>Rad>St
ar
DC: DS>SS
>Shared>NS>Rad>Star

AC:
DS>NS>Shared>SS>
Rad>Star
DC:
DS>NS>Shared>SS>
Star>Rad

Not studied
First bus of
same feeder
Transmission
Not studied Drop: DC > AC
Rise: AC > DC
Drop: DC > AC
Rise: AC > DC
Design
Not studied Designs:
Small differences for
stable cases. Rad and
star unstable.

AC:
Star>Shared>NS>DS>SS>Rad
DC:
Star>Shared>DS>NS>SS>Rad
Bus closest
to the fault
and closer
to the
substation
Transmission
Not studied Not studied Drop: DC > AC
Rise: AC > DC
Design
Not studied Not studied Small differences
Bus closest
to the fault,
further
away from
the
substation
Transmission
Not studied Not studied Drop: DC > AC
Rise: AC > DC
Design
Not studied Not studied AC: SS>DS>NS>Shared
DC: DS>SS>Shared>NS
Rad and star unstable
Table 11 - Comparison of the voltage deviation for dynamic simulations. A>B means that the maximum deviation from
the steady state value is larger for A than for B, i.e. the response is worse.
The table shows that there is a clear difference between using HVDC Light transmission and AC cable
transmission to shore. The AC/DC layouts consequently give larger voltage drops and lower voltage tops
than the AC/AC layouts.

The voltage response can be explained by looking at the short circuit power at the offshore substation.
Short circuit power, SC MVA, is a normal parameter for measuring the strength at a bus in an AC
network. The definition of short circuit power is given in equation 9-2:
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 9
Discussion

- 84 -


th
ac nom
ac nom fault
Z
V
V I SCMVA
2
,
, = =
9-2

The HVDC Light connection provides a limited short circuit current, upwards limited to 1pu , based on
the converter rating. The short circuit power contribution to the offshore substation is therefore limited to
the nominal pu value of the AC voltage, which is 1 pu. This gives that the maximum short circuit power
delivered by the HVDC Light converter is 1 pu, referred to the converter MVA base. This is 570 MVA for
each converter, giving that the maximum short circuit current from the HVDC Light converter at the
offshore substation is 1140 MVA. Referring this to the base power for the offshore substation, which is
the system base (100 MVA), the maximum short circuit power from the two HVDC Light converters is
11.4 pu based on the system base.

For an AC transmission, there is no such upper limit. By performing an automatic sequence short circuit
fault calculation at the offshore substation bus in PSS/E. it was found that for the different designs, the
short circuit current contribution from the transmission system was 25.64 to 25.91 pu, or 2564MVA to
2591MVA. From Ohms law, one can conclude that, for constant short circuit impedance, the higher the
short circuit current is, the higher the short circuit voltage is. Therefore, all systems using AC transmission
to shore give smaller voltage drops than the systems using HVDC Light transmission.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 10
Conclusion

- 85 -

10 Conclusion
Presently, there is a debate among the offshore wind community regarding the value of redundancy
required in the offshore grid to maximize the energy yield, and the impact it may impose on the capital
costs of the wind farm. The major concern is related to the cost of supplementary subsea cabling, either in
terms of extra length or higher ratings, versus the value of the decreased losses during normal operation
and contingency operation.

In this thesis, six different designs of the electrical grid of a 540 MW offshore wind farm, placed 100km
off the Norwegian coast, have been studied. The six different designs are the radial design, the single
sided ring, the double sided ring, the shared ring, the n-sided ring (with n equal to four in this thesis) and
the star design. In addition, two transmission options to shore have been studied. For both of the
transmission options, two parallel connections to shore were installed in order to provide redundancy. The
first option was to have two parallel AC cable connections to shore, with a rated voltage level of 400 kV.
The second option was to have two parallel HVDC Light connections to shore, rated at 150kV DC. The
main scope has been to study the performance of the different wind farm layouts. The power losses during
normal and contingency running conditions have been found and compared. In addition to this, the
reliability of the systems has been compared by looking at the expected amount of cable failures and the
consequence of any cable fault in the offshore grid. Finally, the dynamic response following a fault has
been studied in order to evaluate which layout is the most robust when it comes to withstanding a fault in
the offshore system.

All simulations have been executed in version 31 of the program PSS/E .The wind farm was modeled full
scale, consisting of 108 wind turbines rated at 5MW. The wind turbines were modeled as doubly fed
induction generators, using the generic wind model that comes with version 31 of PSS/E.

The load flow simulations showed that the differences in power losses were quite large, both when
comparing the six grid designs and when looking at the transmission to shore. The choice of transmission
system had a larger impact on the system losses than the choice of offshore grid design. Nevertheless, the
design also influenced the losses considerably. A connection using two AC cable connections to shore
gave lower total system losses than using two HVDC light connections.

The lowest losses when using AC cables were found at the n-sided ring design. The losses in this design
were found to be 12.6MW, which is 2.33% of the total produced power in the wind farm. The highest
losses were found in the star design. The losses in this design were 22.5MW, which is 4.17% of the total
produced power in the wind farm. When using HVDC Light, the losses were considerably higher: The
lowest losses when using HVDC Light were found at the n-sided bus. The losses in this design were
34.6MW, which is 6.41% of the total produced power in the wind farm. The highest losses were found in
the star design. The losses in this design were 44.2MW, which is 8.19% of the total produced power in the
wind farm.

An additional load flow simulation was run, studying the offshore wind farm under worst case running
conditions. Under these conditions, it is assumed that a fault has led to the disconnection of one of the
cables connecting the innermost wind turbine of one of the feeders to the offshore substation. The radial
design and the star design provide no redundancy, so for these systems the entire feeder where the cable is
disconnected is unable to deliver any power to the offshore substation. The lowest losses were still found
for the n-sided design, using AC cable transmission to shore. These losses are now 12.9MW, which is
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 10
Conclusion

- 86 -

2.39% of the total produced power in the wind farm. The highest losses are again found at the star design,
using HVCD Light transmission to shore. For this system design, the losses are 86.2MW, which is 15.96%
of the total produced power in the wind farm with all cables in function.

It must be noted that in the AC/AC layouts, there was a need for fixed reactors at a total size of
approximately 1600 MVAr for all cases. In addition, an SVC of minimum 175 MVArs had to be installed
in order to fulfill the grid codes regarding power factor control of the wind farm output power. This
confirms what was written in the introduction, where it was stated that long AC cable connections to shore
can lead to the need of very large and expensive reactive power compensators.

In terms of reliability, the looped designs had the superiorly smallest consequences of losing a cable.
Also, for the n-sided and double sided designs, the expected number of faults was only slightly higher than
for the radial design, since the amount of extra cabling is small.

To decide what effect the offshore grid design and transmission system design had on the dynamic
response of the offshore wind turbines following a fault, the voltage response and rotor speed response of
selected generators in the wind farm was studied when subject to a three phase short circuit fault, lasting
150ms. Three fault points were studied. The first fault point was in the connection to shore, the second
was in the cable connecting the innermost wind turbine in a feeder to the offshore substation, and the third
was in the cable connecting the two outermost wind turbines of a feeder. The faults gave varying results
when comparing the different grid designs. It is difficult to draw any conclusions as regarding which
offshore grid design gives the best offshore dynamic response, as this seems to be more dependent on
where the fault occurs than what grid design was chosen. When comparing the transmission systems
though, it was clear that for nearly all offshore grid designs and all fault types, the speed deviations were
larger when using HVDC Light transmission compared to using AC cables. Also, for all simulations, the
HVDC Light transmission to shore gave increased voltage drops offshore. This is due to the low short
circuit power capability of the HVDC connection compared to the AC cable connection. Thus, the results
indicate that HVDC Light has a negative influence on the offshore stability when compared to AC cables.
Thomas Haugsten Hansen
Offshore wind farm layouts

Chapter 11
Further work

- 87 -

11 Further work
This thesis is the end of the work that started in the project work executed during the fall of 2008. After
working with the subject of offshore wind power and the connection of offshore wind power to the grid, it
has become increasingly clear that the work that has been executed in this thesis only covers a small part
of the challenges attached to the offshore wind power field.

This thesis does not focus on the control systems of the wind turbines. To compare the dynamic behavior
of different layouts without optimizing the control system for each design might lead to different results
than what could be achieved with better tuned and better designed control systems. A natural way to pick
up and continue the work with this thesis is to develop the control systems of the wind turbines and the
SVC. Also, several studies indicate that an offshore DC grid, as presented in chapter 3.4, might be
competitive to an offshore AC grid in terms of profitability. This has not been studied in this thesis, and
should be further looked into.

When using an offshore DC grid, there are several fields that should be investigated. The design of the
power electronics in this system, covering the DC generator, an offshore DC transformer and a medium or
high frequency transformer for the wind turbine are suggestions to what the

The wind farm economy should be investigated in order to better compare the different layouts. When
given the power production and power losses of the wind farm, it would be very interesting to dig into the
issues of reliability and to conduct a feasibility study, where the profitability of the different layouts is
compared.

Furthermore, in the offshore wind power community the idea of having an offshore multi terminal DC
(MTDC) grid in the North Sea, connecting the countries surrounding it to a cluster of offshore wind farms
and also to each other has been discussed. The first offshore MTDC grid is planned at Kriegers Flak,
connecting Sweden, Denmark and Germany together and to the offshore wind farm. The effects of a
similar offshore grid on the Norwegian power system should be studied. Also, the effect on the power
market should be investigated.

Also, the idea of connecting the MTDC-grid to one or more oil platforms, or to connect smaller wind
farms to the platforms in isolated power systems, should be investigated as this would provide
environmentally friendly energy to the oil platforms, the idea of connecting the MTDC-grid to one or
more oil platforms, or to connect smaller wind farms to the platforms in isolated power systems should be
investigated as this would provide environmentally friendly energy to an industry which is not exactly
known for its environmentally friendly image

Thomas Haugsten Hansen
Offshore wind farm layouts


Bibliography

- 88 -

Bibliography

ABB. (2008). Retrieved september 23, 2008, from ABB HVDC - Power T&D solutions:
http://www.abb.com/hvdc
ABB. (2008). It's time to connect - Technical description of HVDC Light technology. ABB.
ABB Power technologies AB. (2009). User guide for the PSS/E implementation of the HVDC Light
detailed model Version 1.1.7. Ludvika: ABB.
Ackermann, T. (2005). Wind power in power systems. John Wiley & Sons, Ltd.
Barberis, N., Todorovic, J., & Ackermann, T. (2005). Loss evaluation of HVAC and HVDC transmission
solutions for large offshore wind farms. Science Direct.
Elkington, K. (2009). Modelling and control of doubly fed induction generators in power systems
(Licentiate thesis). Stockholm, Sweden: KTH, electrical engineering.
Ericsson nkt. (n.d.). http://www.encv.co.uk/Products/Datasheets.aspx. Retrieved May 30, 2009, from
Ericsson nkt cables Venture Ltd: http://www.encv.co.uk/Products/Datasheets.aspx
Gardner, P., Craig, L. M., & Smith, G. J. Electrical Systems for Offshore Wind Farms. Glasgow, UK:
Garrad Hassan & Partners.
Gjengedal, T. (2009, May 13). Telephone conversation.
Haugsten Hansen, T. (2008). Offshore wind power - power system studies. Project work at Department of
Electrical Engineering, NTNU.
Hubert, C. I. (2002). Electric Machines - Theory, Operations, Applications, Adjustment, and Control.
Prentice Hall.
Kundur, P. (1994). Power System Stability and Control. McGraw-Hill.
Kundur, P., Paserba, J., Ajjarapu, V., Andersson, G., Bose, A., Canizares, C., et al. (2003). Definition and
classification of power system stability. IEEE.
Lundberg, S. (2006). Wind Farm Configuration and Energy Efficiency Studies - Series DC versus AC
Layouts. Department of Energy and Environment, Chalmers University of Technology.
Machowski, J., Bialek, J. W., & Bumby, J. R. (1998). Power system dynamics and stability. John Wiley &
Sons Ltd.
Martander, O. (2002). DC grids for wind farms. Department of electric power engineering, Chalmers
university of technology.
Mohan, & Undeland. (2006). Power Electronics. John Wiley & Sons.
NordPool. (2009). All Spot - NordPool. Retrieved May 18, 2009, from ASA NordPool web site:
http://www.nordpool.com/system/flags/elspot/area/all/
Quinonez-Varela, G., Ault, G., Anaya-Lara, O., & McDonald, J. (2007, June). Electrical collector system
options for large offshore wind farms. IET Renewable Power Generation, Vol.1, No.2 .
RePower Systems AG. (n.d.). RePower Systems AG: Wind Turbines. Retrieved May 13, 2009, from
http://www.repower.de/index.php?id=12&L=1
Schtte, T., Gustavsson, B., & Strm, M. (2001). The use of low frequency AC for Offshore Wind Power.
Proceedings of the second international workshop on transmission networks for offshore wind farms.
Edited by Thomas Ackermann, Royal institute of technology.
Siemens PTD. (2008, april 14). Power point presentation: HVDC plus; technology benefits &
applications.
Siemens PTI. (2007). PSSE users manual - Program application guide - Volume II. Siemens Power
Transmission & Distribution, Inc., Power Technologies International.
Thomas Haugsten Hansen
Offshore wind farm layouts


Bibliography

- 89 -

Statnett SF. (n.d.). Dokumenter - Statnett - Samfunnskonomi.pdf. Retrieved May 15, 2009, from Statnett
SF (Norwegian): http://www.statnett.no/Prosjekter/Kabel-til-Nederland-NorNed/Dokumenter/
Statnett SF. (2008). Funksjonskrav i kraftsystemet (Norwegian). Statnett SF.
Tavner, P. (2009, april 23). Reliability & availability of wind turbine electrical & electronic components
(power point presentation). Wind power to the grid - EPE Wind Energy Chapter - 2nd seminar,
Stockholm, Sweden.
Vestas Wind systems AS. (n.d.). Vestas V90 - 3.0 MW - information brochure. Retrieved april 26, 2009,
from Vestas web page: http://www.vestas.com/en/wind-power-solutions/wind-turbines/3.0-mw
www.hornsrev.dk. (n.d.). Horns rev offshore wind farm. Retrieved May 8, 2009, from
http://www.hornsrev.dk/index.en.html

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

I

Appendix 1: Load flow reports

Radial design
ACAC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************
SYSTEM SWING BUS SUMMARY

X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X MW MVAR MVABASE
1 PCC 400.00 1 1 15.4 0.0 1000.0

220 BUSES 110 PLANTS 110 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 221 BRANCHES 109 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 555.4 -11.4 555.4 -11.4
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1611.8 0.0 1611.8
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1769.6 0.0 1715.5

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 7.06 75.41 0.0 0.0 1762.8
33.0 216 8.30 70.96 0.0 0.0 6.7
TOTAL 221 15.35 146.38 0.0 0.0 1769.6

ACAC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************
SYSTEM SWING BUS SUMMARY

X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME X BASKV # X-- NAME X MW MVAR MVABASE
1 PCC 400.00 1 1 59.1 0.0 1000.0

202 BUSES 101 PLANTS 101 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 203 BRANCHES 100 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 554.1 -24.4 554.1 -24.4
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1611.8 0.0 1611.8
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1769.0 0.0 1714.9

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

II

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 6.42 67.35 0.0 0.0 1762.8
33.0 198 7.65 65.42 0.0 0.0 6.2
TOTAL 203 14.07 132.76 0.0 0.0 1769.0

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

III

ACDC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************
SYSTEM SWING BUS SUMMARY

X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME X BASKV # X-- NAME X MW MVAR MVABASE
1 PCC 300.00 1 1 37.4 -32.3 1000.0
112 OFFSH HVDC 195.00 1 1 -264.2 -78.6 570.0
212 OFFSH HVDC 2195.00 1 1 -264.2 -78.6 570.0

224 BUSES 113 PLANTS 113 MACHINES 4 FIXED SHUNTS 0 SWITCHED
SHUNTS
1 LOADS 222 BRANCHES 113 TRANSFORMERS 0 DC LINES 0 FACTS
DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 551.6 -223.1 551.6 -223.1
TO CONSTANT POWER LOAD 540.0 -44.9 540.0 -44.9
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 6.7 0.0 6.4

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
300. 0 3 0.00 25.11 0.0 0.0 0.0
195.0 2 0.00 27.76 0.0 0.0 0.0
132.0 1 3.24 46.26 0.0 0.0 0.0
33.0 216 8.35 71.37 0.0 0.0 6.7
TOTAL 222 11.59 170.50 0.0 0.0 6.7

ACDC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************
SYSTEM SWING BUS SUMMARY

X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 300.00 1 1 79.8 10.5 1000.0
112 OFFSH HVDC 195.00 1 1 -242.3 -79.7 570.0
212 OFFSH HVDC 2195.00 1 1 -242.3 -79.7 570.0

206 BUSES 104 PLANTS 104 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 204 BRANCHES 104 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X

MW MVAR MW MVAR
FROM GENERATION 550.4 -199.3 550.4 -199.3
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

IV

FROM LINE CHARGING 0.0 6.1 0.0 5.9

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
300.0 3 0.00 21.05 0.0 0.0 0.0
195.0 2 0.00 23.34 0.0 0.0 0.0
132.0 1 2.72 38.91 0.0 0.0 0.0
33.0 198 7.66 65.50 0.0 0.0 6.1
TOTAL 204 10.38 148.80 0.0 0.0 6.1

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

V

Single sided loop design
ACAC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 400.00 1 1 13.3 0.0 1000.0

220 BUSES 110 PLANTS 110 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 233 BRANCHES 109 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 553.3 -34.4 553.3 -34.4
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1611.5 0.0 1611.5
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1783.5 0.0 1728.8

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 7.09 75.80 0.0 0.0 1762.8
33.0 228 6.25 61.76 0.0 0.0 20.6
TOTAL 233 13.34 137.56 0.0 0.0 1783.5
ACAC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY

X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 400.00 1 1 13.8 0.0 1000.0

220 BUSES 110 PLANTS 110 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 232 BRANCHES 109 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 553.8 -32.2 553.8 -32.2
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1611.5 0.0 1611.5
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1782.9 0.0 1728.3

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 7.08 75.72 0.0 0.0 1762.8
33.0 227 6.68 63.55 0.0 0.0 20.1
TOTAL 232 13.76 139.27 0.0 0.0 1782.9

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

VI

ACDC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X

BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 300.00 1 1 35.4 12.7 1000.0
112 OFFSH HVDC 195.00 1 1 -265.2 -68.4 570.0
212 OFFSH HVDC 2195.00 1 1 -265.2 -68.4 570.0

224 BUSES 113 PLANTS 113 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 234 BRANCHES 113 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 549.5 -200.2 549.5 -200.2
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 20.1 0.0 19.7

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
300.0 2 0.00 12.65 0.0 0.0 0.0
195.0 3 0.00 40.72 0.0 0.0 0.0
33.0 229 9.54 108.56 0.0 0.0 20.1
TOTAL 234 9.54 161.93 0.0 0.0 20.1

ACDC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 300.00 1 1 35.8 12.6 1000.0
112 OFFSH HVDC 195.00 1 1 -265.0 -68.4 570.0
212 OFFSH HVDC 2195.00 1 1 -265.0 -68.4 570.0

224 BUSES 113 PLANTS 113 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 233 BRANCHES 113 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 550.0 -198.1 550.0 -198.1
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 19.7 0.0 19.2

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

VII

300.0 2 0.00 12.63 0.0 0.0 0.0
195.0 3 0.00 40.66 0.0 0.0 0.0
33.0 228 9.96 110.27 0.0 0.0 19.7
TOTAL 233 9.96 163.56 0.0 0.0 19.7

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

VIII

Double sided loop design
ACAC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 GRID 400.00 1 1 13.0 0.0 1000.0

220 BUSES 110 PLANTS 110 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 227 BRANCHES 109 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 553.0 -19.9 553.0 -19.9
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1611.5 0.0 1611.5
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1768.1 0.0 1714.1

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 7.09 75.87 0.0 0.0 1762.8
33.0 222 5.90 60.84 0.0 0.0 5.3
TOTAL 227 12.99 136.71 0.0 0.0 1768.1

ACAC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 GRID 400.00 1 1 13.6 0.0 1000.0

220 BUSES 110 PLANTS 110 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 226 BRANCHES 109 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 553.6 -16.9 553.6 -16.9
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1611.5 0.0 1611.5
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1767.9 0.0 1713.9

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 7.08 75.75 0.0 0.0 1762.8
33.0 221 6.51 63.71 0.0 0.0 5.0
TOTAL 226 13.59 139.47 0.0 0.0 1767.9

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

IX

ACDC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 GRID 300.00 1 1 35.0 12.7 1000.0
112 OFFSH HVDC 195.00 1 1 -265.4 -68.4 570.0
212 OFFSH HVDC 2195.00 1 1 -265.4 -68.4 570.0

224 BUSES 113 PLANTS 113 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 228 BRANCHES 113 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 549.3 -185.4 549.3 -185.4
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 5.2 0.0 5.0

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
300.0 1 0.00 0.00 0.0 0.0 0.0
195.0 3 0.00 39.40 0.0 0.0 0.0
132.0 2 3.28 60.88 0.0 0.0 0.0
33.0 222 5.97 61.52 0.0 0.0 5.2
TOTAL 228 9.25 161.80 0.0 0.0 5.2

ACDC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 GRID 300.00 1 1 35.6 12.6 1000.0
112 OFFSH HVDC 195.00 1 1 -265.1 -68.4 570.0
212 OFFSH HVDC 2195.00 1 1 -265.1 -68.4 570.0

224 BUSES 113 PLANTS 113 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 227 BRANCHES 113 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 549.9 -182.5 549.9 -182.5
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 4.9 0.0 4.8

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

X

300.0 1 0.00 0.00 0.0 0.0 0.0
195.0 3 0.00 39.31 0.0 0.0 0.0
132.0 2 3.27 60.74 0.0 0.0 0.0
33.0 221 6.58 64.40 0.0 0.0 4.9
TOTAL 227 9.85 164.45 0.0 0.0 4.9


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XI

Shared loop design
ACAC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 400.00 1 1 14.5 0.0 1000.0

235 BUSES 110 PLANTS 110 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 248 BRANCHES 109 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 554.5 -17.8 554.5 -17.8
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1611.7 0.0 1611.7
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1772.4 0.0 1718.1

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 7.07 75.57 0.0 0.0 1762.8
33.0 231 7.47 67.33 0.0 0.0 9.5
0.0 12 0.00 0.00 0.0 0.0 0.0
TOTAL 248 14.54 142.90 0.0 0.0 1772.4

ACAC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 400.00 1 1 15.0 0.0 1000.0

235 BUSES 110 PLANTS 110 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 247 BRANCHES 109 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 555.0 -15.1 555.0 -15.1
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1611.7 0.0 1611.7
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1771.8 0.0 1717.6

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 7.06 75.47 0.0 0.0 1762.8
33.0 230 7.98 69.60 0.0 0.0 9.0
0.0 12 0.00 0.00 0.0 0.0 0.0
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XII

TOTAL 247 15.04 145.07 0.0 0.0 1771.8

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XIII

ACDC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 GRID 300.00 1 1 36.6 12.6 1000.0
112 OFFSH HVDC 195.00 1 1 -264.6 -68.5 570.0
212 OFFSH HVDC 2195.00 1 1 -264.6 -68.5 570.0

239 BUSES 113 PLANTS 113 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 249 BRANCHES 113 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 550.8 -183.8 550.8 -183.8
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 9.3 0.0 9.1

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
300.0 3 0.00 25.19 0.0 0.0 0.0
195.0 2 0.00 27.94 0.0 0.0 0.0
132.0 1 3.26 46.56 0.0 0.0 0.0
33.0 231 7.54 67.87 0.0 0.0 9.3
0.0 12 0.00 0.00 0.0 0.0 0.0
TOTAL 249 10.80 167.56 0.0 0.0 9.3

ACDC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 GRID 300.00 1 1 37.2 12.6 1000.0
112 OFFSH HVDC 195.00 1 1 -264.3 -68.5 570.0
212 OFFSH HVDC 2195.00 1 1 -264.3 -68.5 570.0

239 BUSES 113 PLANTS 113 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 246 BRANCHES 113 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 551.3 -181.1 551.3 -181.1
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 8.8 0.0 8.6

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XIV

LEVEL BRANCHES MW MVAR MW MVAR MVAR
300.0 3 0.00 25.13 0.0 0.0 0.0
195.0 2 0.00 27.88 0.0 0.0 0.0
132.0 1 3.25 46.47 0.0 0.0 0.0
33.0 229 8.07 70.26 0.0 0.0 8.8
0.0 11 0.00 0.00 0.0 0.0 0.0
TOTAL 246 11.33 169.74 0.0 0.0 8.8

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XV

N-sided loop design (N=4)
ACAC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 400.00 1 1 12.6 0.0 1000.0

220 BUSES 110 PLANTS 110 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 230 BRANCHES 109 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 552.6 -29.3 552.6 -29.3
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1611.4 0.0 1611.4
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1784.8 0.0 1730.1

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 7.10 75.95 0.0 0.0 1762.8
33.0 225 5.46 68.09 0.0 0.0 21.9
TOTAL 230 12.56 144.04 0.0 0.0 1784.8

ACAC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************
SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 400.00 1 1 12.9 0.0 1000.0

220 BUSES 110 PLANTS 110 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 229 BRANCHES 109 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 552.9 -25.2 552.9 -25.2
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1611.4 0.0 1611.4
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1783.8 0.0 1729.2

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 7.09 75.89 0.0 0.0 1762.8
33.0 224 5.81 71.31 0.0 0.0 21.0
TOTAL 229 12.91 147.20 0.0 0.0 1783.8

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XVI

ACDC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 300.00 1 1 34.6 12.7 1000.0
112 OFFSH HVDC 195.00 1 1 -265.6 -68.4 570.0
212 OFFSH HVDC 2195.00 1 1 -265.6 -68.4 570.0

224 BUSES 113 PLANTS 113 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 231 BRANCHES 113 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 548.8 -194.9 548.8 -194.9
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 21.5 0.0 21.1

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
300.0 2 0.00 12.69 0.0 0.0 0.0
195.0 3 0.00 40.85 0.0 0.0 0.0
132.0 1 3.28 46.92 0.0 0.0 0.0
33.0 225 5.48 68.16 0.0 0.0 21.5
TOTAL 231 8.76 168.61 0.0 0.0 21.5

ACDC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 300.00 1 1 35.3 12.7 1000.0
112 OFFSH HVDC 195.00 1 1 -265.4 -68.4 570.0
212 OFFSH HVDC 2195.00 1 1 -265.4 -68.4 570.0

224 BUSES 113 PLANTS 113 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 230 BRANCHES 113 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 549.1 -190.9 549.1 -190.9
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 20.7 0.0 20.1

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XVII

300.0 2 0.00 12.67 0.0 0.0 0.0
195.0 3 0.00 40.76 0.0 0.0 0.0
132.0 1 3.28 46.86 0.0 0.0 0.0
33.0 224 5.83 71.42 0.0 0.0 20.7
TOTAL 230 9.11 171.71 0.0 0.0 20.7

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XVIII

Star design
ACAC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 400.00 1 1 22.5 0.0 1000.0

232 BUSES 110 PLANTS 110 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 233 BRANCHES 121 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 562.5 30.6 562.5 30.6
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1612.8 0.0 1612.8
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1770.2 0.0 1716.1

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 6.95 74.05 0.0 0.0 1762.8
33.0 15 4.65 29.42 0.0 0.0 6.2
11.0 213 10.86 84.56 0.0 0.0 1.1
TOTAL 233 22.45 188.04 0.0 0.0 1770.2

ACAC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 400.00 1 1 65.6 0.0 1000.0

213 BUSES 101 PLANTS 101 MACHINES 2 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 214 BRANCHES 111 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 560.6 14.5 560.6 14.5
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 1612.8 0.0 1612.8
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 1769.6 0.0 1715.5

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
400.0 5 6.33 66.20 0.0 0.0 1762.8
33.0 14 4.35 28.11 0.0 0.0 5.7
11.0 195 9.94 76.95 0.0 0.0 1.0
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XIX

TOTAL 214 20.62 171.27 0.0 0.0 1769.6

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XX

ACDC, normal running conditions:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 300.00 1 1 44.2 12.2 1000.0
122 OFFSH HVDC 195.00 1 1 -260.6 -78.7 570.0
222 OFFSH HVDC 2195.00 1 1 -260.6 -78.7 570.0

236 BUSES 113 PLANTS 113 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 234 BRANCHES 125 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 558.7 -138.4 558.7 -138.4
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 7.4 0.0 7.0

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
300.0 3 0.00 24.43 0.0 0.0 0.0
195.0 2 0.00 27.01 0.0 0.0 0.0
132.0 1 3.15 45.02 0.0 0.0 0.0
33.0 15 4.68 29.61 0.0 0.0 6.2
11.0 213 10.90 84.90 0.0 0.0 1.2
TOTAL 234 18.73 210.96 0.0 0.0 7.4

ACDC, worst case:

************************ SUMMARY FOR COMPLETE SYSTEM ************************

SYSTEM SWING BUS SUMMARY
X------ SWING BUS ------X X---- AREA -----X X---- ZONE -----X
BUS# X-- NAME --X BASKV # X-- NAME --X # X-- NAME --X MW MVAR MVABASE
1 PCC 300.00 1 1 86.2 10.2 1000.0
122 OFFSH HVDC 195.00 1 1 -239.0 -79.8 570.0
222 OFFSH HVDC 2195.00 1 1 -239.0 -79.8 570.0

217 BUSES 104 PLANTS 104 MACHINES 4 FIXED SHUNTS 0 SWITCHED SHUNTS
1 LOADS 215 BRANCHES 115 TRANSFORMERS 0 DC LINES 0 FACTS DEVICES

X------ ACTUAL ------X X----- NOMINAL ------X
MW MVAR MW MVAR
FROM GENERATION 556.9 -162.6 556.9 -162.6
TO CONSTANT POWER LOAD 540.0 0.0 540.0 0.0
TO CONSTANT CURRENT 0.0 0.0 0.0 0.0
TO CONSTANT ADMITTANCE 0.0 0.0 0.0 0.0
TO BUS SHUNT 0.0 -342.0 0.0 -342.0
TO FACTS DEVICE SHUNT 0.0 0.0 0.0 0.0
TO LINE SHUNT 0.0 0.0 0.0 0.0
FROM LINE CHARGING 0.0 6.7 0.0 6.4

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 1
Load flow reports

XXI

VOLTAGE X----- LOSSES -----X X-- LINE SHUNTS --X CHARGING
LEVEL BRANCHES MW MVAR MW MVAR MVAR
300.0 3 0.00 20.47 0.0 0.0 0.0
195.0 2 0.00 22.71 0.0 0.0 0.0
132.0 1 2.65 37.86 0.0 0.0 0.0
33.0 14 4.36 28.16 0.0 0.0 5.6
11.0 195 9.93 76.91 0.0 0.0 1.1
TOTAL 215 16.95 186.12 0.0 0.0 6.7

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXII

Appendix 2: Load flow figures

PU-curves used to set up the simulations:

AC transmission 2 lines
RAD SS DS Shared NS Star
P U P U P U P U P U P U
525,1 1,023 528,2 1,025 528,3 1,024 526,3 1,024 529,1 1,024 511,2 1,02
526,1 1,025 529,4 1,027 529,7 1,027 528,1 1,027 530,3 1,027 515,4 1,023
527,2 1,027 530,1 1,029 530,4 1,029 528,9 1,029 531 1,029 518,3 1,026
528,0 1,029 530,6 1,031 530,9 1,031 529,5 1,031 531,5 1,031 520,5 1,029
528,7 1,031 531 1,033 531,2 1,033 529,9 1,033 531,8 1,033 521,5 1,031
529,1 1,033 531,2 1,035 531,4 1,035 530,1 1,035 532 1,035 521,9 1,032
529,3 1,035 531,2 1,037 531,4 1,037 530,1 1,037 532 1,037 522,6 1,035
529,2 1,038 531 1,039 531,3 1,039 529,9 1,039 531,9 1,039 522,6 1,037


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXIII







1,02
1,022
1,024
1,026
1,028
1,03
1,032
1,034
1,036
1,038
1,04
525 526 527 528 529 530
U
_
1
1
0
P_110
Radial ACAC
U
Umax
1,02
1,022
1,024
1,026
1,028
1,03
1,032
1,034
1,036
1,038
1,04
528 528,5 529 529,5 530 530,5 531 531,5
U
_
1
1
0
P_110
SS ACAC
U
Umax
1,02
1,022
1,024
1,026
1,028
1,03
1,032
1,034
1,036
1,038
1,04
528 528,5 529 529,5 530 530,5 531 531,5
U
_
1
1
0
P_110
DS ACAC
U
Umax
1,02
1,022
1,024
1,026
1,028
1,03
1,032
1,034
1,036
1,038
1,04
526 527 528 529 530 531
U
_
1
1
0
P_110
Shared ACAC
U
Umax
1,02
1,022
1,024
1,026
1,028
1,03
1,032
1,034
1,036
1,038
1,04
528,5 529 529,5 530 530,5 531 531,5 532 532,5
U
_
1
1
0
P_110
NS ACAC
U
Umax
1,02
1,022
1,024
1,026
1,028
1,03
1,032
1,034
1,036
1,038
510 512 514 516 518 520 522 524
U
_
1
1
0
P_110
Star ACAC
U
Umax
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXIV

DC, 2 Transmission lines
RAD SS DS Shared NS Star
P U P U P U P U P U P U
522 0,962 527,1 0,97 527,6 0,97 525,5 0,97 528,1 0,97 514,7 0,97
524,4 0,97 528,9 0,98 529,3 0,98 527,5 0,98 529,8 0,98 518,2 0,98
526,5 0,98 530 0,99 529,9 0,985 528,2 0,985 530,4 0,985 519,5 0,985
527,9 0,99 530,3 0,995 530,4 0,99 528,7 0,99 530,8 0,99 520,4 0,99
528,2 0,995 530,5 0,998 530,6 0,994 529 0,994 531,1 0,994 520,7 0,9925
528,4 1 530,5 1 530,7 0,998 529,2 0,998 531,2 0,997 521 0,995
528,4 1,002 530,5 1,0025 530,7 1 529,2 1 531,3 1 521,2 0,9975
528,4 1,004 530,5 1,005 530,7 1,003 529,2 1,005 531,3 1,005 521,3 1
528,4 1,005 530,4 1,0075 530,7 1,006 529,2 1,006 531,2 1,007 521,3 1,001


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXV








0,955
0,96
0,965
0,97
0,975
0,98
0,985
0,99
0,995
1
1,005
1,01
520 522 524 526 528 530
U_110
P_110
RAD ACDC
U
0,965
0,97
0,975
0,98
0,985
0,99
0,995
1
1,005
1,01
526 527 528 529 530 531
U_110
P_110
SS ACDC
U
0,965
0,97
0,975
0,98
0,985
0,99
0,995
1
1,005
1,01
527 528 529 530 531
U_110
DS ACDC
U
0,965
0,97
0,975
0,98
0,985
0,99
0,995
1
1,005
1,01
525 526 527 528 529 530
U_110
Shared ACDC
U
0,965
0,97
0,975
0,98
0,985
0,99
0,995
1
1,005
1,01
527 528 529 530 531 532
U_110
NS ACDC
U
0,965
0,97
0,975
0,98
0,985
0,99
0,995
1
1,005
514 516 518 520 522
U_110
Star ACDC
U
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXVI

Single line diagrams
In the following, single line diagrams showing the load flow situation for all cases are shown:
AC/AC:

Transmission, both lines operating

Transmission, one line operating


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXVII

Radial design, entire system

Radial design, transmission, OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXVIII

Radial design, feeder OK



Radial design, transmission, worst case


Radial design, feeder, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXIX

Single sided design, entire system

Single sided design, transmission, OK


Single sided design, feeder OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXX

Single sided design, transmission, worst case


Single sided design, feeder worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXXI

Double sided design, entire system

Double sided design, transmission, OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXXII

Double sided design, feeder OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXXIII

Double sided design, transmission, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXXIV

Double sided design, feeder worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXXV

Shared ring design, entire system

Shared ring design, transmission, OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXXVI

Shared ring design, feeder OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXXVII

Shared ring design, transmission, worst case


Shared ring design, feeder, worst case

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXXVIII



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XXXIX

N-sided ring, entire system

N-sided ring, transmission, OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XL

N-sided ring, feeder OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XLI

N-sided ring, transmission, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XLII

N-sided ring, feeder, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XLIII

Star design, entire system

Star design, transmission, OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XLIV

Star design, feeder OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XLV

Star design, transmission, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XLVI

Star design, feeder, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XLVII

AC/DC

Transmission, both HVDC connections in function

Transmission, one HVDC connection in function


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XLVIII

Radial design, entire system

Radial design, transmission, OK


Radial design, feeder OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

XLIX

Radial design, transmission, worst case


Radial design, feeder, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

L

Single sided design, entire system

Single sided design, transmission, OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LI

Single sided design, feeder OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LII

Single sided design, transmission, worst case


Single sided design, feeder, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LIII

Double sided design, entire system

Double sided design, transmission, OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LIV

Double sided design, feeder OK


Double sided design, transmission, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LV

Double sided design, feeder, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LVI

Shared ring design, entire system

Shared ring design, transmission, OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LVII

Shared ring design, feeder OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LVIII

Shared ring design, transmission, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LIX

Shared ring design, feeder, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LX

N-sided design, entire system

N-sided design, transmission, OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LXI

N-sided design, feeder OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LXII

N-sided design, transmission, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LXIII

N-sided design, feeder, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LXIV

Star design, entire system

Star design, transmission, OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LXV

Star design, feeder OK



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LXVI

Star design, transmission, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 2
Load flow figures

LXVII

Star design, feeder, worst case



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 3
ASCC reports

LXVIII

Appendix 3: Automatic sequence short circuit reports
Appplying a fault to the offshore substation bus, these are the ASCC-reports from PSS/E for the AC/AC systems, showing
the fault current contributions from wind turbines and the transmission to shore:

Radial design
PSS/E SHORT CIRCUIT OUTPUT
WED, JUL 15 2009 11:46 .
HOME BUS IS 1000. OFFSH SUB 33.000.

. *** FAULTED BUS IS: 1000 [OFFSH SUB 33.000] *** . 0 LEVELS AWAY

AT BUS 1000 [OFFSH SUB 33.000] AREA 1 (PU) V+: / 0.0000/ 0.00

THEV. R, X, X/R: POSITIVE 0.00272 0.03262 11.999

T H R E E P H A S E F A U L T
X--------- FROM ----------X AREA CKT I/Z /I+/ AN(I+) /Z+/ AN(Z+) APP X/R
2 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
11 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
20 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
29 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
38 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
47 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
56 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
65 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
74 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
83 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
92 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
101 [WTHV 33.000] 1 1 PU/PU 0.6274 -39.60 0.0799 77.72 4.594
110 [TRANS OFFSH 400.00] 1 1 PU/PU 25.9139 -90.52 0.0158 86.00 14.286
TOTAL FAULT CURRENT (P.U.) 31.2126 -79.73

Single sided design
PSS/E SHORT CIRCUIT OUTPUT WED, JUL 15 2009 11:50 .HOME BUS IS 1000.OFFSH SUB 33.000.
. *** FAULTED BUS IS: 1000 [OFFSH SUB 33.000] *** . 0 LEVELS AWAY .
AT BUS 1000 [OFFSH SUB 33.000] AREA 1 (PU) V+: / 0.0000/ 0.00
THEV. R, X, X/R: POSITIVE 0.00301 0.03283 10.899

T H R E E P H A S E F A U L T
X--------- FROM ----------X AREA CKT I/Z /I+/ AN(I+) /Z+/ AN(Z+) APP X/R
2 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
10 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
11 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
19 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
20 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
28 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
29 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
37 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
38 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
46 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
47 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
55 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
56 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
64 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
65 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
73 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
74 [WTHV 33.000] 1 1 PU/PU 0.3762 -39.56 0.0799 77.72 4.594
82 [WTHV 33.000] 1 1 PU/PU 0.2508 -38.92 0.1358 76.97 4.321
83 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
91 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
92 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
100 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
101 [WTHV 33.000] 1 1 PU/PU 0.3760 -39.82 0.0799 77.72 4.594
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 3
ASCC reports

LXIX

109 [WTHV 33.000] 1 1 PU/PU 0.2507 -39.20 0.1358 76.97 4.321
110 [TRANS OFFSH 400.00] 1 1 PU/PU 25.6420 -90.01 0.0158 86.00 14.286
TOTAL FAULT CURRENT (P.U.) 30.9779 -79.22

Double sided design
. PSS/E SHORT CIRCUIT OUTPUT WED, JUL 15 2009 11:52 .HOME BUS IS 1000.
. .OFFSH SUB 48.000.
. . .
. *** FAULTED BUS IS: 1000 [OFFSH SUB 48.000] *** . 0 LEVELS AWAY .

AT BUS 1000 [OFFSH SUB 48.000] AREA 1 (PU) V+: / 0.0000/ 0.00

THEV. R, X, X/R: POSITIVE 0.00299 0.03271 10.926

T H R E E P H A S E F A U L T
X--------- FROM ----------X AREA CKT I/Z /I+/ AN(I+) /Z+/ AN(Z+) APP X/R
2 [WTHV 48.000] 1 1 PU/PU 1.2286 -38.35 0.0422 78.32 4.838
20 [WTHV 48.000] 1 1 PU/PU 0.6385 -40.62 0.0422 78.32 4.838
29 [WTHV 48.000] 1 1 PU/PU 0.6385 -40.62 0.0422 78.32 4.838
38 [WTHV 48.000] 1 1 PU/PU 0.6385 -40.62 0.0422 78.32 4.838
47 [WTHV 48.000] 1 1 PU/PU 0.6385 -40.62 0.0422 78.32 4.838
56 [WTHV 48.000] 1 1 PU/PU 0.6385 -40.62 0.0422 78.32 4.838
65 [WTHV 48.000] 1 1 PU/PU 0.6385 -40.62 0.0422 78.32 4.838
74 [WTHV 48.000] 1 1 PU/PU 0.6385 -40.62 0.0422 78.32 4.838
83 [WTHV 48.000] 1 1 PU/PU 0.6385 -40.62 0.0422 78.32 4.838
92 [WTHV 48.000] 1 1 PU/PU 0.6385 -40.62 0.0422 78.32 4.838
101 [WTHV 48.000] 1 1 PU/PU 0.6385 -40.62 0.0422 78.32 4.838
110 [TRANS OFFSH 400.00] 1 1 PU/PU 25.6419 -90.01 0.0158 86.00 14.286
TOTAL FAULT CURRENT (P.U.) 31.1076 -79.24

Shared ring system
. PSS/E SHORT CIRCUIT OUTPUT WED, JUL 15 2009 11:52 .HOME BUS IS 1000.
. .OFFSH SUB 33.000.
. . .
. *** FAULTED BUS IS: 1000 [OFFSH SUB 33.000] *** . 0 LEVELS AWAY .

AT BUS 1000 [OFFSH SUB 33.000] AREA 1 (PU) V+: / 0.0000/ 0.00

THEV. R, X, X/R: POSITIVE 0.00302 0.03280 10.847

T H R E E P H A S E F A U L T
X--------- FROM ----------X AREA CKT I/Z /I+/ AN(I+) /Z+/ AN(Z+) APP X/R
2 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
11 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
20 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
29 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
38 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
47 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
56 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
65 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
74 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
83 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
92 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
101 [WTHV 33.000] 1 1 PU/PU 0.5312 -39.75 0.0799 77.72 4.594
110 [TRANS OFFSH 400.00] 1 1 PU/PU 25.6418 -90.00 0.0158 86.00 14.286
22301 [ _2 ] 1 1 PU/PU 0.3911 -39.71 0.1439 77.71 4.592
22801 [ _7 ] 1 1 PU/PU 0.3911 -39.71 0.1439 77.71 4.592
23301 [ _< ] 1 1 PU/PU 0.3911 -39.71 0.1439 77.71 4.592
TOTAL FAULT CURRENT (P.U.) 31.0150 -79.22

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 3
ASCC reports

LXX

N-sided design (n=4)
. PSS/E SHORT CIRCUIT OUTPUT WED, JUL 15 2009 11:54 .HOME BUS IS 1000.
. .OFFSH SUB 33.000.
. . .
. *** FAULTED BUS IS: 1000 [OFFSH SUB 33.000] *** . 0 LEVELS AWAY .

AT BUS 1000 [OFFSH SUB 33.000] AREA 1 (PU) V+: / 0.0000/ 0.00

THEV. R, X, X/R: POSITIVE 0.00302 0.03293 10.916

T H R E E P H A S E F A U L T
X--------- FROM ----------X AREA CKT I/Z /I+/ AN(I+) /Z+/ AN(Z+) APP X/R
2 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
11 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
20 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
29 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
38 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
47 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
56 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
65 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
74 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
83 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
92 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
101 [WTHV 33.000] 1 1 PU/PU 0.6219 -39.17 0.0647 83.88 9.331
110 [TRANS OFFSH 400.00] 1 1 PU/PU 25.6422 -90.02 0.0158 86.00 14.286
TOTAL FAULT CURRENT (P.U.) 30.9003 -79.23
Star design

. PSS/E SHORT CIRCUIT OUTPUT WED, JUL 15 2009 12:14 .HOME BUS IS 1000.
. .COLLECTOR 33.000.
. . .
. *** FAULTED BUS IS: 1000 [COLLECTOR 33.000] *** . 0 LEVELS AWAY .

AT BUS 1000 [COLLECTOR 33.000] AREA 1 (PU) V+: / 0.0000/ 0.00

THEV. R, X, X/R: POSITIVE 0.00289 0.03320 11.496

T H R E E P H A S E F A U L T
X--------- FROM ----------X AREA CKT I/Z /I+/ AN(I+) /Z+/ AN(Z+) APP X/R
110 [TRANS OFFSH 400.00] 1 1 PU/PU 25.9115 -90.44 0.0158 86.00 14.286
205 [STAR1 33KV_133.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
206 [STAR2 33KV_133.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
207 [STAR3 33KV_133.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
208 [STAR4 33KV_133.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
305 [STAR1 33KV_233.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
306 [STAR2 33KV_233.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
307 [STAR3 33KV_233.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
308 [STAR4 33KV_233.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
405 [STAR1 33KV 33.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
406 [STAR2 33KV 33.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
407 [STAR3 33KV 33.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
408 [STAR4 33KV 33.000] 1 1 PU/PU 0.5944 -36.48 0.0799 77.72 4.594
TOTAL FAULT CURRENT (P.U.) 30.6559 -79.60
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXI

Appendix 4: Dynamic models
This appendix describes all the dynamic models used in this thesis. Block diagrams and parameters are given for all
models, except from the ABB HVDC Light user model.

Wind turbine model
The figure underneath shows the principal scheme of the WT3 generic wind model in PSS/E.



All issues of making an equivalent of a wind farm to be modeled are up to the user. The model provides the possibility to
lump several turbines into one equivalent representation. The user should make a decision on how many original units that
can be lumped into one equivalent machine presented in the load flow case. For n lumped machines, the machine rating
M
BASE
of the original machine must be multiplied by n. It is also up to the user to take care of the adequate equivalent of
the wind farm feeders, collectors, and step up transformers.

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXII


Generator/converter model (WTRG1)

Parameters:
CONs ICONs
Xeq 0.8000 No of lumped turbines 1
PLL gain 30.0000
PLL integrator gain 0.0000
PLL min limit 0.1000
Turbine MW rating 5.0000


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXIII

Electrical control model(WT3E1)

Reactive power control system:

Torque/current control:


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXIV

Parameters:
CONs ICONs
Tfv - V-regulator filter 0.150 Remote bus # 110
Kpv - V-regulator proportional gain 18.00
VARFLG: =0 Const. Q ctrl, =1 reactive power ctrl,
= -1 const. pf ctrl 1
Kiv - V-regulator integrator gain 5.000 VLTFLG: =1 Closed loop terminal voltage control
AC: 0
DC: 1
Xc - line drop compensation
reactance 0.050 From bus - interconnection transformer 301
Tfp - T-regulator filter 0.050 To bus - interconnection transformer 3
Kpp - T-regulator proportional gain 3.000 Id - interconnection transformer '1'
Kip - T-regulator integrator gain 0.600
PMX - T-regulator max limit 1.120
PMN - T-regulator min limit 0.100
QMX - V-regulator max limit 0.350
QMN - V-regulator min limit -0.436
IPMAX - Max active current limit 1.100
TRV - V-sensor 0.050
RPMX - maximum Pordr derivative 0.450
RPMN - minimum Pordr derivative 0.450
T_POWER - Power filter time
constant 5.000
KQi - MVAR/Volt gain 0.050
VMINCL 0.900
VMAXCL 1.200
Kqv - Volt/MVAR gain 40.00
XIQmin - min. limit of (Vterm -
Eq'cmd) 0.5000
XIQmax - max. limit of (Vterm -
Eq'cmd) 0.4000
Tv - Lag time constant in WindVar
controller 0.0500
Tp - Pelec filter in fast PF controller 0.0500
Fn - A portion of on-line wind
turbines 1.0000
Shaft speed at Pmin, pu 0.6900
Shaft speed at 20% rated power, pu 0.7800
Shaft speed at 40% rated power, pu 0.9800
Shaft speed at 60% rated power, pu 1.1200
Shaft speed at 80% rated power, pu 0.7400
Shaft speed at 100% rated power, pu 1.2000


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXV

Wind turbine model (WT3T1)




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXVI

Parameters:
CONs ICONs
Vw - Initial wind speed, pu of rated wind speed 1.2500
H - Total inertia constant, MW*sec/MVA 4.9500
DAMP - Machine damping factor, pu P/pu speed 0.0000
Kaero - Aerodynamic gain factor 0.0070
Theta2 - Blade pitch at twice rated wind speed, deg. 21.9800
Htfac - Turbine inertia fraction (Hturb/H) 0.8750
Freq1 - First shaft torsional resonant frequency, Hz 1.8000
DSHAFT - Shaft Damping factor, pu P/pu speed 1.5000


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXVII

Pitch control model (WT3P1)



Parameters:
CONs ICONs
Tp - Time constant of the output lag (sec) 0.3000
Kpp - Proportional gain of PI regulator(pu) 150.0000
Kip - Integrator gain of PI regulator (pu) 25.0000
Kpc - Proportional gain of the compensator(pu) 3.0000
Kic - Integrator gain of the compensator (pu) 30.0000
TetaMin - Lower pitch angle limit (degrees) 0.0000
TetaMax - Upper pitch angle limit (degrees) 27.0000
RTetaMax - Upper pitch angle rate limit (deg/sec) 10.0000
PMX - Power reference (pu) 1.0000

Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXVIII

SVC-model (CSVGN5)

Parameters:
CONs ICONs
TS1 0.0000 IB, remote bus number 0
VE MAX 0.1500
TS2 0.1000
TS3 (>0) 5.0000
TS4 0.0000
TS5 0.0000
KSVS 400.0
KSD 0.0000
BMAX 1.0000
B'MAX 1.0000
B'MIN -0.5000
B MIN -0.5000
TS6 (>0) 0.0500
DV 0.1500


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXIX

Onshore generator model

Salient pole generator model (GENSAL):


Parameters:
CONs ICONs
T'do (> 0) 7.0000
T''do (> 0) 0.0400
T''qo (> 0) 0.2500
Inertia H 20.0000
Speed Damping D 0.0000
Xd 2.3000
Xq 1.4000
X'd 0.2768
X''d = X''q 0.2000
X1 0.1200
S(1.0) 0.1000
S(1.2) 0.3000


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXX

Generator exciter model (SCRX):



Parameters:
CONs ICONs
TA/TB 0.1000
TB (> 0) 10.0000
K 200.0000
TE 0.0500
EMIN 0.0000
EMAX 4.0000
CSWITCH (0=bus fed, 1=solid fed) 0.0000
rc/rfd 0.0000


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 4
Dynamic models

LXXXI

Generator governor model (HYGOV)


Parameters:
CONs ICONs
R, Permanent Droop 0.0500
r, Temporary Droop 0.4000
Tr (> 0) Governor Time Constant 8.0000
Tf (> 0) Filter Time Constant 0.0500
Tg (> 0) Servo Time Constant 0.2000
VELM, Gate Velocity Limit 0.1000
GMAX, Maximum Gate Limit 2.0000
GMIN, Minimum Gate Limit -2.0000
TW (> 0) Water Time Constant 2.0000
At, Turbine Gain 1.1000
Dturb, Turbine Damping 1.7500
qNL, No Load Flow 0.1000
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

LXXXII

Appendix 5: Dynamic plots
This appendix shows all plots of P,Q, and U at the investigated buses in this thesis. Bus 201 (6701 for the AC/DC star
design) is the first bus of the feeder where no fault occurs. Bus 3801 (5601 for the star design) is the first bus of the feeder
where the fault occurs. Bus 4501 (5601 for the star designs) is the second outermost bus in the feeder where the fault
occurs. Bus 4601 (6501 for the star designs) is the outermost bus in the feeder where the fault occurs. For all plots, the
color code is as follows:

Green: Active power [pu on system base]
Blue: Reactive power [pu on system base]
Light blue: Generator speed [pu on system base]
Purple: Voltage [pu on system base]

RADIAL DESIGN

AC/AC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

LXXXIII

FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

LXXXIV

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

LXXXV



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

LXXXVI

AC/DC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

LXXXVII

FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

LXXXVIII

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

LXXXIX




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

XC

SINGLE SIDED DESIGN

AC/AC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

XCI

FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION





Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

XCII

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

XCIII




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

XCIV

AC/DC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

XCV


FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION





Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

XCVI

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

XCVII



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

XCVIII

DOUBLE SIDED DESIGN

AC/AC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

XCIX

FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION





Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

C

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CI



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CII

AC/DC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CIII


FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CIV

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CV




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CVI

SHARED RING DESIGN

AC/AC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CVII

FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION





Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CVIII

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CIX



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CX

AC/DC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXI

FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXII

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXIII



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXIV

N-SIDED RING DESIGN

AC/AC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXV

FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION





Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXVI

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXVII



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXVIII

AC/DC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXIX

FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXX

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXI



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXII

STAR DESIGN

AC/AC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT


Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXIII

FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION





Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXIV

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXV






Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXVI

AC/DC TRANSMISSION

FAULT TYPE 1: TRANSMISSION FAULT



Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXVII

FAULT TYPE 2: FAULT BETWEEN INNERMOST TURBINE OF FEEDER AND SUBSTATION




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXVIII

FAULT TYPE 3: FAULT IN THE OUTERMOST CABLE IN THE FEEDER




Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXIX





Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXX

COMPARISON, MINIMUM AND MAXIMUM VALUES

Fault 1. Bus 201. , P, U minimum and maximum values
-0,0568 -0,0569 -0,0571 -0,0569 -0,0569
-0,0550
-0,0646 -0,0637 -0,0645 -0,0642 -0,0644
-0,0658
0,0609 0,0611 0,0611 0,0609 0,0616 0,0606
0,0690 0,0682 0,0691 0,0686 0,0688
0,0724
-0,0800
-0,0600
-0,0400
-0,0200
0,0000
0,0200
0,0400
0,0600
0,0800
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
-0,0485 -0,0486 -0,0487 -0,0485 -0,0489
-0,0475
-0,0489 -0,0490 -0,0491 -0,0490 -0,0492
-0,0484
0,0053 0,0053 0,0054 0,0053 0,0053 0,0053
0,0115 0,0113 0,0111
0,0116 0,0117
0,0103
-0,0600
-0,0500
-0,0400
-0,0300
-0,0200
-0,0100
0,0000
0,0100
0,0200
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
P deviation
-72,8 %
-74,8 % -74,8 %
-73,7 % -74,0 %
-63,3 %
-77,5 %
-78,4 % -78,7 %
-77,8 % -77,7 %
-69,8 %
7,4 % 7,6 % 7,5 % 7,4 % 7,5 % 7,1 %
2,4 % 2,7 %
5,1 %
2,8 % 3,1 %
1,6 %
-90,0 %
-80,0 %
-70,0 %
-60,0 %
-50,0 %
-40,0 %
-30,0 %
-20,0 %
-10,0 %
0,0 %
10,0 %
20,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXXI


Fault 2. Bus 201 , P, U minimum and maximum values
-0,0114
-0,0169
-0,0270
-0,0209
-0,0242
-0,0119
-0,0100
-0,0158
-0,0296
-0,0216
-0,0248
-0,0113
0,0104
0,0169
0,0275
0,0211
0,0247
0,0110
0,0097
0,0163
0,0307
0,0222
0,0256
0,0109
-0,0400
-0,0300
-0,0200
-0,0100
0,0000
0,0100
0,0200
0,0300
0,0400
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
-0,0093
-0,0136
-0,0209
-0,0164
-0,0188
-0,0091
-0,0105
-0,0145
-0,0226
-0,0169
-0,0198
-0,0107
0,0043
0,0046
0,0058
0,0051
0,0057
0,0052 0,0053
0,0033
0,0046
0,0037
0,0045
0,0064
-0,0250
-0,0200
-0,0150
-0,0100
-0,0050
0,0000
0,0050
0,0100
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
P deviation
-27,2 %
-35,0 %
-48,6 %
-40,7 %
-43,7 %
-26,2 % -26,1 %
-33,2 %
-50,6 %
-40,5 %
-43,6 %
-26,8 %
5,5 %
5,0 %
6,8 %
4,8 % 4,7 % 4,9 %
0,2 %
-1,1 %
3,6 %
0,5 % 0,6 % 0,5 %
-60,0 %
-50,0 %
-40,0 %
-30,0 %
-20,0 %
-10,0 %
0,0 %
10,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXXII




Fault 2 3801/5601 , P, U max/min
0,0000
-0,0583 -0,0585 -0,0583 -0,0587
0,0000 0,0000
-0,0589
-0,0606
-0,0595 -0,0599
0,0000 0,0000
0,0624 0,0625 0,0623 0,0628
0,0000 0,0000
0,0646
0,0660
0,0650 0,0653
0,0000
-0,0800
-0,0600
-0,0400
-0,0200
0,0000
0,0200
0,0400
0,0600
0,0800
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
0,0000
-0,0493 -0,0492 -0,0492 -0,0493
0,0000 0,0000
-0,0496 -0,0496 -0,0496 -0,0496
0,0000 0,0000
0,0106 0,0109 0,0109
0,0098
0,0000 0,0000
0,0055
0,0068
0,0058 0,0058
0,0000
-0,0600
-0,0500
-0,0400
-0,0300
-0,0200
-0,0100
0,0000
0,0100
0,0200
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
P deviation
-86,5 % -86,7 % -86,1 % -86,5 % -86,6 %
-74,8 %
-89,7 % -89,5 % -89,5 % -89,5 % -89,5 %
-80,8 %
27,9 %
4,7 %
6,9 %
4,7 % 4,2 %
32,7 %
14,6 %
-0,1 %
5,1 %
1,4 % 1,2 %
11,6 %
-100,0 %
-80,0 %
-60,0 %
-40,0 %
-20,0 %
0,0 %
20,0 %
40,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXXIII



Fault 3. Bus 3801/5601 , P, U minimum and maximum values
-0,0352
-0,0395
-0,0431
-0,0450 -0,0442
-0,0577
-0,0374
-0,0409
-0,0464 -0,0472
-0,0462
-0,0602
0,0365
0,0411
0,0448
0,0468 0,0460
0,0618
0,0388
0,0432
0,0491
0,0504
0,0490
0,0651
-0,0800
-0,0600
-0,0400
-0,0200
0,0000
0,0200
0,0400
0,0600
0,0800
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
-0,0291
-0,0324
-0,0352
-0,0364
-0,0359
-0,0492
-0,0309
-0,0337
-0,0371
-0,0380
-0,0372
-0,0496
0,0070 0,0074
0,0082
0,0077 0,0077
0,0110
0,0060
0,0047
0,0056 0,0051 0,0052
0,0070
-0,0600
-0,0500
-0,0400
-0,0300
-0,0200
-0,0100
0,0000
0,0100
0,0200
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
P deviation
-59,7 %
-65,5 %
-70,4 %
-73,2 %
-71,3 %
-86,0 %
-63,0 %
-68,1 %
-74,7 %
-76,7 %
-74,6 %
-89,9 %
3,6 %
5,0 %
6,9 %
4,7 % 4,8 %
1,3 %
0,5 %
-0,6 %
3,7 %
0,7 % 0,7 % 0,2 %
-100,0 %
-80,0 %
-60,0 %
-40,0 %
-20,0 %
0,0 %
20,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXXIV




Fault 3. Bus 4501/5601. , P, U minimum and maximum values
-0,0573 -0,0579 -0,0580 -0,0586 -0,0586 -0,0577
-0,0590 -0,0585
-0,0601 -0,0599 -0,0597 -0,0602
0,0615 0,0621 0,0621 0,0626 0,0626 0,0618
0,0638 0,0640
0,0654 0,0654 0,0651 0,0651
-0,0800
-0,0600
-0,0400
-0,0200
0,0000
0,0200
0,0400
0,0600
0,0800
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
-0,0492 -0,0493 -0,0492 -0,0492 -0,0493 -0,0492 -0,0496 -0,0496 -0,0496 -0,0496 -0,0496 -0,0496
0,0114 0,0113
0,0120
0,0113 0,0108 0,0110
0,0064
0,0055
0,0067
0,0060 0,0058
0,0070
-0,0600
-0,0500
-0,0400
-0,0300
-0,0200
-0,0100
0,0000
0,0100
0,0200
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
P deviation
-86,5 % -86,7 % -86,1 % -86,5 % -86,6 % -86,0 %
-89,7 % -89,5 % -89,5 % -89,6 % -89,5 % -89,9 %
3,4 %
4,9 %
6,9 %
4,6 % 4,7 %
1,3 % 0,7 %
-0,4 %
3,9 %
0,8 % 0,9 % 0,2 %
-100,0 %
-80,0 %
-60,0 %
-40,0 %
-20,0 %
0,0 %
20,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)
Thomas Haugsten Hansen
Offshore wind farm layouts

Appendix 5
Dynamic plots

CXXXV




Fault 3. Bus 4601/6501. , P, U minimum and maximum values

0,0000
-0,0552
-0,0562
-0,0523
-0,0548
0,0000 0,0000
-0,0560
-0,0586
-0,0544
-0,0563
0,0000 0,0000
0,0589
0,0599
0,0552
0,0582
0,0000 0,0000
0,0609
0,0636
0,0586
0,0611
0,0000
-0,0800
-0,0600
-0,0400
-0,0200
0,0000
0,0200
0,0400
0,0600
0,0800
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
Speed deviation
0,0000
-0,0466
-0,0472
-0,0432
-0,0456
0,0000 0,0000
-0,0474
-0,0481
-0,0445
-0,0465
0,0000 0,0000
0,0110 0,0114
0,0096 0,0098
0,0000 0,0000
0,0064 0,0065
0,0056 0,0056
0,0000
-0,0600
-0,0500
-0,0400
-0,0300
-0,0200
-0,0100
0,0000
0,0100
0,0200
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
P deviation
-86,4 %
-84,5 % -84,3 %
-81,3 %
-83,3 %
-85,4 %
-89,7 %
-88,5 % -89,0 %
-85,7 %
-87,5 %
-89,5 %
27,8 %
5,1 %
6,9 %
4,6 % 4,7 %
31,3 %
14,1 %
-0,5 %
4,0 %
0,8 % 0,9 %
10,4 %
-100,0 %
-80,0 %
-60,0 %
-40,0 %
-20,0 %
0,0 %
20,0 %
40,0 %
RAD AC SS AC DS AC SH AC NS AC ST AC RAD DC SS DC DS DC SH DC NS DC ST DC
U drop/rise (% of original value)